This chapter examines mechanisms of aid coordination management used in the four cases studied, namely Education (SWAp), Health (SWiM and TB Sub-sector Program), Local Governance (Seila), and Public Finance (TCAP). After a brief overview of the four programs, institutional arrangements to manage aid coordination are reviewed to discuss the nature and characteristics of the respective cases. This chapter then looks at the process of aid coordination management at each stage of the program cycle: preparation and planning (including policy and strategy formulation), implementation, and monitoring and evaluation.

The Programs studied have a relatively short history (3-5 years), with the exception of the Seila Program (7 years) and the Tuberculosis Sub-Sector (10 years) (see Table 2-1 for a summary of the programs). All the programs are ongoing as of 2003, although TCAP is scheduled to be completed in 2003. The Health and Education Sectors have been major recipients of external assistance totaling more than $100 million, or more than 20% of total external assistance, in 2001.¹⁴ More than 10 donors are involved in the Education SWAp and Health SWiM programs, and the numbers appear to be increasing. It should be noted that NGOs are not included in the number of donors in Table 2-1, due to the unavailability of data, although NGO funds are included in the amount of external assistance for Health and Seila. As NGO assistance has accounted for more than 10% of external assistance in the past (with the exception of public finance), the need for aid coordination is even greater than these figures indicate.

A brief overview of each program is provided below as a background.

The Ministry of Education, Youth and Sport (MOEYS) introduced the concept of a sector-wide approach for national education reforms in 1999.¹⁵ This reflected unsatisfactory progress prior to 1999 during which donor assistance had been substantial (total $244 million in 1994-99), and yet had made only a limited impact.¹⁶ This disappointing performance was attributed to the lack of an integrated policy framework and the proliferation of discreet and ill-coordinated donor-funded projects. The SWAp was started to ensure that all external assistance projects would support a single sector policy and expenditure program developed under government leadership, and that a common approach to planning, implementation and monitoring evaluation would be adopted. The MOEYS has been developing a SWAp process in which a wide range of stakeholders could participate, including an Education Strategic Plan (ESP) and a rolling Education Sector Support Program (ESSP). A number of aid modalities, including budget support and project-support, have been accepted to ensure the participation of a broad range of donors. The MOEYS, donors and NGOs jointly reviewed the performance of the education sector in 2002 and 2003.

The Ministry of Health proposed introducing a program in the health sector in early-1999 through the Health Sector Reform Project (HSR III). This program was later renamed Sector Wide Management (SWiM). The concept of SWiM is similar to that of a sector-wide approach, but a key difference is that SWiM explicitly does not require budget support through the pooling of funds for the implementation of projects. Conducting a series of consultative processes in which a wide range of stakeholders participated, the government and its external partners jointly developed a Health Sector Strategic Plan 2003-2007 (HSP) and a rolling Medium Term Expenditure Framework (MTEF) in August 2002. The HSP was recently developed further into a 5-year Implementation Plan. The first joint sector review was held in April 2003.

The National Tuberculosis Control Program (NTP) is one of the four health sub-sector programs under the overall national health system in Cambodia. The National Center for Tuberculosis and Leprosy Control (CENAT) has been managing this Program since 1980. The NTP supports TB control activities with the provision of technical advice, supplies, training and outreach activities. The CENAT recently developed the "National Health Strategic Plan for Tuberculosis Control 2001-2005," a new 5-year policy and strategy for TB control, in broad consultation with stakeholders. The government and all development partners have endorsed this Plan under which subsequent planning and implementation of activities have been carried out. All donor assistance continues to be provided through project support, but is well coordinated under the agreed Plan.

Seila started in 1996 as a national program with core support from four donors. Seila’s overall goal is to institute decentralized systems and strategies for poverty alleviation and good governance. Its specific objectives are to foster local development and poverty alleviation, build the capacity of provincial and commune authorities for managing sustainable development, and to generate lessons for development of national policies for deconcentration and decentralization. Seila developed unique decentralized systems for planning, programming, financing, implementation, monitoring and reporting for both provincial and commune levels. These Seila systems include components such as organizations, management instruments and processes that can be applied in any local administration and governments in Cambodia. The election of Commune/Sangkat (C/S) Councils in February 2002 established the first elected administrative bodies at the sub-national level, and the government at the national and provincial levels assumed responsibility for support to all C/S Councils in the country. Seila has developed two funds, the Provincial Investment Fund and Commune/Sangkat Fund, which are financed by national budget and external assistance, to deliver infrastructure and services to provinces and communes, respectively. Since 1996, Seila has gradually increased the number of provinces it supports, and expanded to cover all 24 provinces/municipalities and all 1,621 C/S Councils in 2003.

Public Finance TCAP

TCAP is a comprehensive technical assistance program covering economic and public financial policy. It was developed under the lead of IMF and jointly with four other donors in 2000. The Program’s objectives are to strengthen the institutional and human resource capacities of six key Departments and agencies: Tax, Customs, Budget and Treasury at MEF, the National Bank of Cambodia and the National Institute of Statistics. Program activities include the provision of advisors and experts for policy formulation, organization of seminars, workshops and study tours for capacity building, and the provision of materials and equipment for implementation. Under the Program, the government has prepared a number of policy papers, research papers and legal documents for economic and fiscal policy reforms that are supported by a Poverty Reduction and Growth Facility (PRGF) of IMF and a Structural Adjustment Credit (SAC) of the World Bank. For human resource development, the program organized a series of seminars, trainings and study tours. The program is scheduled to be completed by the the first quarter of 2004.

2. Institutional arrangements for aid coordination

Institutional arrangements are a critical component to managing programs and coordinating external assistance. Table 2-2 summarizes the institutional arrangements of the cases studied. There appear to be five key arrangements that have played critical roles in managing programs and coordinating external assistance: (i) overall coordination among all stakeholders; (ii) coordination within the government; (iii) coordination among donor agencies; (iv) coordination among NGOs; and (v) a core team or secretariat that implements programs and coordinates external assistance. These institutional arrangements are discussed in turn.

(i) Overall coordination among all stakeholders

All cases studied have an institution that is mandated to facilitate information sharing, dialogue and coordination, and/or make collective decisions for all stakeholders involved, including the government, donors and NGOs. This could be either an institution that already existed before the program (CoCom for Health SWiM and ICC for Health TB Sub-Sector), or a newly created one (the Government/Donor/NGO Consultative Meeting in Education, the National Seila Forum and District Integration Workshop in Seila, and the Tripartite Review Meeting in TCAP). A common feature of those institutions is that they serve as a forum where all stakeholders meet regularly to exchange views and opinions, and make collective decisions about the program. The frequency and intensity of the meetings appear to vary, depending on what roles and responsibilities are assigned to the institution.

(ii) Coordination within the government

The process of aid coordination can be broken down into three stages, similar to a program cycle: (i) preparation and planning, (ii) implementation, and (iii) monitoring, evaluation and review. The modality and intensity of aid coordination vary considerably, depending on the stages studied. Table 2-3 summarizes key activities at each stage for the cases studied.

In all the cases studied, the government undertook preparation of programs with the support of donors and with broad participation of stakeholders involved. Some key points are highlighted below.

In Education SWAp, Health SWiM and the Health TB Sub-Sector, the government developed a sector (or sub-sector) policy and strategy with joint efforts of stakeholders, including external partners. This was an important step toward enhanced aid coordination. The following activities were critical for the preparation of the plan and strategy:

Financing modality of donor assistance is a key issue at the implementation stage because it affects government ownership, capacity, and the administrative burden on the government.

In the cases studied, a large part of donor funding is provided through project support, and the government and some donors are making efforts to implement projects in closely coordinated ways. For instance, some projects or components under a program are co-financed or cost-shared using trust fund arrangements. MOH recently launched a Health Sector Support Project (HSSP), which is co-financed by ADB, World Bank and DFID, to implement the SWiM process. Seila’s core component is cost-shared by Sida, DFID and UNDP with the use of trust fund arrangements. TCAP also has a component cost-shared by UNDP, DFID and Netherlands, and the other components are co-financed by IMF and ADB.

There are also some other examples of close collaboration among different projects in the health sector, in particular among national programs receiving substantial donor assistance.¹⁹ Implementing different activities in a harmonized way and conducting joint activities under a common policy framework are among the various collaboration schemes utilized. For example, in the TB sub-sector, many activities have been supported in a coordinated way by a number of donors including JICA, World Bank, WHO, and NGOs. One of the collaborative activities was the dissemination of the National TB Prevalence Survey (2001), which was supported by JICA and the World Bank.

Among the cases studied, Education SWAp is the only case in which a ministry receives direct budget support from donor agencies (ADB and EU). This budget support consists of 18 percent of total external assistance in the education sector in 2003, providing non-wage recurrent costs under the government’s Priority Action Programs (PAPs).²⁰ The funds pass through the recurrent budget of the government via the treasury system of the government.

Some donor representatives and experts interviewed argued that budget support should play a more important role, but many donors were concerned with the potential fiduciary risks in the case of Cambodia because of the government’s weak capacity for managing public finance. Indeed, a recent joint study by World Bank and ADB reported that the government’s capacity to manage public finance remains weak and that there is wide scope for improvement.²¹ Also, a donor representative pointed out that too high a concentration of external assistance in direct budget support might cause a major decline in the national budget and a subsequent drop in service delivery if donor funding had to be suspended for whatever reasons. In addition, a recent problem of the delay in PAP disbursements in Education suggests that enhancing the performance of PAPs would require substantial reforms in the treasury system of the government as well as enhancement of fiduciary management ability of line ministries at both national and sub-national levels. Implementing these measures will require substantial time and resources. Under the circumstances, it is likely that project-type support will continue to be the main financing modality, and that budget support will be used by only some donors where necessary and appropriate, at least in the short to medium term.

Donor assistance has been provided largely through project-type support under the cases studied. This implies that aid procedures during implementation (such as procurement, disbursement, financial management, accounting, auditing and reporting) differ between projects, depending on the policies and procedures set out by supporting donors. Under project-type support, the government typically needs to maintain two accounting and reporting systems to satisfy both government and donor requirements for each project, which imposes substantial transaction costs to the government. Few examples of harmonizing aid procedures at implementation have been found in the cases studied, except for the trust funds arrangements in the components of TCAP and Seila discussed above. In these cases, participating donors have adopted a common procedure for the use of the trust funds (UNDP national execution modality) agreed between the government and participating donors, thereby reducing transaction costs to the government to some extent.

Salary supplements for government officials pose a major challenge to the government and its development partners in Cambodia. The problems arising from salary supplements in Cambodia have been long known, but few coordinated actions have been taken to address them. The low pay in the civil service coupled with the practice of uncoordinated salary supplements in donor projects distorts incentives in the civil service in a number of ways. First is accountability. Civil servants who work for donor projects are expected to be accountable to donors; therefore their regular work in the government is often compromised, and the government loses access to already scarce public sector talent. Second is the inequality between civil servants who receive salary supplements and those who do not. The amount the former group had received totaled nearly six million US dollars annually, whereas the rest of the civil servants receive none.²² It is not uncommon that junior officials with salary supplements receive higher pay than senior officials with greater responsibilities. Third, the selection of officials who work for donor funded projects is usually not transparent, demoralizing those who have not been selected. Fourth, competition over competent personnel for donor projects puts pressure on donors to pay even higher salary supplements in order to secure counterparts for their own projects, further exacerbating the situation.

One other major consequence of salary supplements is its adverse effect on sustainability of civil service reform to address low pay problems. A World Bank study on the experience of supporting civil service reforms concluded that the use of salary supplements in donor activities as an interim solution for low pay problem "does not provide enduring answers to fundamental problems of civil service incentives; indeed, they ultimately undermine the likelihood of devising ultimate solutions."²³ The study further noted that "the insidious aspect of this problem is that in many circumstances the offenders are the donors."²⁴

The estimation of salary supplements, or more broadly the provision of financial incentives to government officials, is often a difficult task, as they are in many cases hidden under different activities within project budgets. The available evidences suggest, however, that donors’ projects provide financial incentives in a variety of forms, called per diems, allowances, fees and so on, which are associated with specific activities under donor-funded projects, such as research, field visits, travel, training, translation, and participation in workshops and seminars.²⁵ There are also cases in which donors provide salary supplements for government officials at project implementation units who take leave of absence during the project implementation period.

A comprehensive mapping of salary supplements practices in Cambodia is beyond the scope of this study, and shall be undertaken in the capacity development study under the Government-Donor Partnership Working Group. However, some concrete examples were found in the cases studied. In the Health SWiM, salary supplements were provided for the Core Team during the period of the Health Sector Strategic Plan (HSP) development. The Team members took leave of absence from their MOH positions during that period, and received salary supplements for their work from WHO, UNICEF and GTZ. The Seila staff consists of seconded government officials from line ministries, all of whom are paid salary supplements. Seila programs are managed by 30 officials at the national level (Seila Task Force Secretariat) and around 1,480 officials at the provincial and district levels. The provincial and district level staffs receive salary supplements of $80 and $40 per month, respectively. It was reported that salary supplements provide incentives for them to do their work properly and deliver outputs as intended. Under TCAP, the cost-sharing contributions are pooled and used for salary supplements as well as many other activities such as trainings, seminars and workshops, study tours and administration costs. The government officials who work as national experts for TCAP receive salary supplements of $180 per month, which is the UN standard in Cambodia.

Aid coordination has been actively taking place at the monitoring, evaluation and review stage for all the cases studied. Joint sector or program reviews by the government and donors are carried out in Education SWAp, Health SWiM, Seila core component, and TCAP. In addition, donors supervise projects or sub-components of a program on a regular basis and their reports are shared with other participating donors.

There are signs that aid coordination may proceed further at the monitoring and evaluation and review stage. For instance, GTZ indicated that in the future its project review in the health sector will be integrated into the annual joint health sector review under SWiM. UNFPA intends to take similar action for its review.