UNITED NATIONS CAPITAL DEVELOPME

Document Sample
UNITED NATIONS CAPITAL DEVELOPME Powered By Docstoc
					        UNITED NATIONS CAPITAL DEVELOPMENT FUND



               CAMBODIAN LOCAL DEVELOPMENT FUND
                           CMB/97/CO1



                       FINAL EVALUATION
                          September 1999




Prepared by,

Robert A. Leonard
Dr. Robert Guild
Cristina Mansfield
                                                                                             ii


                             ACKNOWLEDGMENTS




We, the members of the Evaluation Team, would like to express our thanks to all who
made this study possible. Our gratitude goes to Scott Leiper and Joel Charny and all the
CARERE staff who spent many hours interviewing (whom?) and making the appropriate
documents available to us. Our special thanks to the three Provincial Program Officers,
Joanne Morrision of Battambang, Peter Robertson of Banteay Meanchey, and Hans van
Zoggell and his staff, who traveled day after day to the far corners of the provinces with
much enthusiasm and good humor. Also, our thanks to Irene Pietersen for her assistance
and excellent feedback regarding our report.

We would also like to thank the Kingdom of Cambodia for their time and assistance in
this effort.

Our very special thanks go to Laurie Gross at UNCDF in Cambodia, who spent many
hours coordinating our visits, organizing documents, giving us excellent insights for our
report, and generally making our stay enjoyable.
                                                                 iii



              ABBREVIATIONS AND ACRONYMS


AIT      Asian Institute of Technology
CARERE   Cambodia Area Rehabilitation and Regeneration Project
CDC      Commune Development Committee
CASD     Community Action for Social Development
DDC      District Development Committee
DDF      Decentralized Development Fund
DFT      District Facilitation Team
Ex-Com   Executive Committee of the PRDC
IFAD     International Fund for Agriculture and Development
ILO      International Labor Organization
IO       International Organization
IPF      Indicative Planning Figure
KPI      Key Performance Indicator
KoC      Kingdom of Cambodia
LCB      Local Capacity Building
LDF      Local Development Fund
LPP      Local Planning Process
LPS      Local Planning System
M&E      Monitoring and Evaluation
MoF      Ministry of Finance
MoI      Ministry of Interior
MRD      Ministry of Rural Development
NGO      Non-Governmental Organization
PDRD     Provincial Department of Rural Development
PPP      Provincial Planning Process
PRA      Participatory Rural Appraisal
PFT      Provincial Facilitation Team
PRDC     Provincial Rural Development Committee
ProDoc   Project Document
RGC      Royal Government of Cambodia
RRA      Rapid Rural Appraisal
RDS      Rural Development Structure
RGC      Royal Government of Cambodia
TOT      Training of Trainers
TSS      Technical Support Staff
UNCDF    United Nations Capital Development
VDC      Village Development Committee
VDP      Village Development Plan
WFP      World Food Program
                                                                                               iv



                             EXECUTIVE SUMMARY


I      INTRODUCTION

    A. Background

The thirty-year period of war and civil strife has left the country’s infrastructure in
shambles, with secondary roads in constant need of repair, and even main arterial roads
are in need of upgrading. To travel from one part of a district to another in the rainy
season may take as much as two to three hours. Irrigation structures have been poorly
designed and broken down leaving most farmers to depend solely on rainfall for their
crops. In addition, crop loss due to drought and flood has become common. Only a
small percentage of the population has access to reliable water for domestic use. Schools
are in a state of disrepair with leaky roofs and flooded classrooms during the rainy
season, and children have problems getting to school because of the poor condition of
roads.

The greatest concern is the effect on the capacity of human resources and institutions, the
economy, society, culture, religion, and the sense of community. Basic trust between
people in villages has even been affected. Most of the educated people were eliminated,
died of starvation or left the country. The same fate happened to over 75% of the
teachers. These events have had an enormous influence on the process of development as
well, and seasoned development workers in Cambodia are adamant in advising that
projects often take much longer and require much more effort than in other developing
countries.

    B, An Overview

The SEILA/CARERE/LDF project has been called “an experiment in decentralized
planning and financing of participatory rural development.” It is an experiment in
restructuring local government institutions, building their capacity, and furthering civil
society, while at the same time building infrastructure. It is also a participatory planning
process. The Local Development Fund is an important component of the overall project
and provides resources for implementation.

The project is a partnership between government (SEILA), CARERE (a project of
UNDP), and the Local Development Fund supported by UNCDF and other donors.
SEILA is a Royal Government inter-ministerial task force that oversees the project in the
five provinces of Battambang, Banteay Meanchey, Siem Reap, Pursat and Ratanakiri.
These five provincial government administrations implement the project, while CARERE
is the catalyst that propels the program forward through training, technical support, and
management advisory services.
                                                                                              v


In short, the project combines a bottom-up/top-down relationship between local
government at the base and provincial government at the top. The tangible outputs of the
project are rural infrastructure sub-projects, made possible through donor money directed
to the Local Development Fund. A participatory Local Planning Process is the
connecting thread between outputs and the institutions that implement them.

   B. Overall Goals of the Local Development Fund

The overall development objective of the main phase Project Document is as follows:

To alleviate poverty and spread peace in Cambodia by strengthening the bonds linking
society to the structures of the State, and empowering the Cambodian rural population to
become fully participating members in the development process through decentralized
governance.


   C. Immediate Objectives

The project is centered on these five immediate objectives:

1. To develop a participatory, demand-driven process of rural development,
   infrastructure planning and implementation;

2. To fund the construction of priority rural infrastructure projects identified through
   this participatory process;

3. To augment and institutionalize the capacity of the PRDCs;

4. To augment and institutionalize the capacity of the VDCs and CDCs;

5. To establish incentives for increased internal support for rural infrastructure projects
   through financial, in-kind, and technical contributions from central government and
   from local communities, while at the same time developing a sense of “ownership”
   for the local projects.

   C. Attainment of Objectives

As of August 1999, the LDF has supported the establishment of 1,144 locally elected
Village Development Committees (VDCs), 134 Commune Development Committees
(CDCs), and five Provincial Rural Development Committees (PRDCs), five Secretariats
and five Ex-Com Committees, all fully staffed by government civil servants.

Because of the lack of baseline data, it is difficult to measure impact; however, from what
the Evaluation Team (ET) has observed, the project has obtained the above objectives
beyond expectations. Projects are, on the whole, being completed in a satisfactory
manner in less time than planned; villagers are identifying and prioritizing their needs at
                                                                                          vi


the grassroots level; and VDCs and CDCs are turning them into viable projects.
Provincial government, through the newly established PRDC, Executive Committee, and
Secretariat, is establishing these local development bodies and is assisting them in
designing and implementing the projects.

The project has developed a financial management system that is providing transparency,
good tracking, and accountability. In addition, a Decentralized Development Fund has
been established that provides a system for other donors to invest development funds
directly from central government to the provinces. The monitoring and evaluation (M &
E) of the projects is carried out by technical staff of the government who have also
established a bidding process for contractors.

On the whole, local government is different than it was four years ago. Local
government institutions have enthusiastically accepted this bottom-up/top-down style of
development.. They are actively listening to people in the villages and communes and
are engaging them in activities.

There are still many problems, particularly with lack of full village participation in the
process. Some people and groups are not engaged and sometimes are not properly
informed. Maintenance of completed projects is a big problem. SEILA/CARERE know
about this and are actively working to build up a viable system to address this. Capacities
are still low compared to other countries, and planning is still in the beginning stages at
the CDC and VDC level.

To measure success we compare the accomplishments of the present government with the
provincial government of the past. The difference in capacity, participation, inclusion,
communication, planning, and implementation are rather remarkable when one realizes
that the program has only been operating just a little over three years.


II       FINDINGS AND ISSUES

     A. Institutions and Capacity Building

The SEILA/CARERE/LDF project experiment has assisted the newly formed local
government institutions that have dramatically changed the way development is
implemented, and in many respects, the way local government itself operates. These
structures provide the framework for the Local Planning Process to function.

One of the most important institutional factors is the large increase in communication
from the villages through every level of government up to the PRDC. People all through
the system have been taught to brainstorm, to ask questions, and to state opinions.
Higher government bodies are responding through district integration workshops and
through technical and facilitation staff, whose roles are to work with the VDCs and
CDCs. Communication has also been greatly enhanced between sectors through these
                                                                                           vii


various structures. Government staff appears to have a much clearer understanding of its
roles and responsibilities and seems active and motivated with a feeling of “ownership.”

Extensive trainings at all levels have helped this process. In one year alone, 56,245
people (34% women) were trained in 2,200 workshops in such things as planning,
problem solving, finance, water and sanitation, agriculture, management, gender,
governance, health, infrastructure design, and implementation among other things. Most
of this training was for specific job-related functions so they could put the skills learned
into practice. Most important, however, was learning by doing and working with
counterparts in CARERE.

One of the biggest advantages of the new system is that funds stay under regularized
financial management procedures that are subject to internal and external controls. There
are effectively four levels of control, including the Ministry of Economics and Finance
auditors, an internal auditor appointed by Governors (more than one?), external auditors
who may be appointed by CARERE, and the CARERE PSO financial unit.


   B. The Local Planning Process

As of August 1999, SEILA/CARERE/LDF has provided support to 1,144 locally elected
Village Development Committees (VDCs), 134 Commune Development Committees
(CDCs), and five Provincial Rural Development Committees (PRDCs). Each year as the
CDCs and VDCs gain experience in implementing the process, they demonstrate added
maturity in needs assessment, planning, and implementing projects, indicating that
Cambodia has gained acceptance as a viable model for decentralized rural development
with possibilities for replication nationwide.

The rural development structure promotes participation planning to an extent never
before experienced in rural Cambodia. It literally starts at the village and works its way
to the top of provincial government and then back again. To ensure this participation,
dialogue and accountability, the LPP is necessarily complex. The experimental nature of
the program requires that the process be continuously adjusted. Each year, program staff
have reviewed the process and simplified it, using input from all levels and from external
evaluators.

The LPP is also working as a mechanism for planning and coordinating multiple sources
of investment in villages and communes, including LDF, sectoral funds, governmental
department budgets, and outside support from NGOs. Besides interaction between
VDCs, CDCs, and the government, district integration workshops represent a first
opportunity for commune representatives to interact with provincial department officials,
to learn about department policies, and to negotiate the use of resources. The workshops
promote positive interaction between civil society and government and introduce
departments to the concept of responding to public demand.
                                                                                          viii


The LPP is open to other “financial windows” besides SEILA, through the Decentralized
Development Fund. This is supported by recent decisions of the Royal Government,
IFAD, and the World Bank to capitalize the LDF for new projects.

SEILA management is shifting the focus of the program to the commune level. This
makes sense in light of limited resources for individual villages and the expectation of the
upcoming commune elections scheduled for the end of the year 2,000. 1,600 newly-
elected Commune Councils will be created by popular vote, and these bodies will be the
official administrative units of government, as well as development bodies like the CDCs.
The government could request assistance from SEILA/CARERE/LDF in building the
capacity of the new Commune Councils. It is also interested in replicating this model
into other provinces in Cambodia, and would most certainly call on the project staff for
valuable resources.


    C. Delivery of Rural Infrastructure

♦ The sectoral allocation of LDF funds has overwhelmingly favored transportation and
    irrigation works, and to a lesser extent, school buildings and water and sanitation
    projects.
♦   The technical quality of projects in general is good. Of those projects evaluated, over
    80% are acceptable, 18% have problems but can be fixed, and only 2% are
    unacceptable. However, post-completion evaluations are not usually carried out by
    technical support staff.
♦   Competitive bidding was introduced in 1999 and is resulting in lower costs and
    higher quality for projects. Average cost savings compared to estimates range from
    6% to 10%.
♦   The original intent of the allocation process was to use a multicriteria framework that
    included project cost, number of beneficiaries, priority of problems or needs, cost
    effectiveness, time frame, local contribution, and sustainability. This framework has
    proved overly ambitious in the local context and has led to some confusion.
♦   Allocation methods between provinces involve various combinations of poverty
    indicators and population. Some of these indicators are almost certainly colinear and
    therefore do not all need to be included. Others are poor proxies for poverty, such as
    the number of children attending school, which is also strongly affected by
    transportation problems. Using the percentages of female-headed households and
    poor housing may be adequate, but further statistical analysis is required to make a
    recommendation.
♦   The discretion available to PRDCs has led to wide variation in the amounts granted to
    communes, in some cases double or half the average amount over a three year
    funding cycle.
♦   There is widespread agreement that simple criteria for allocation from communes to
    villages work best and are most likely to be applied consistently. Within communes,
    a simple weighting by population may be sufficient.
                                                                                               ix


♦ The current Monitoring and Evaluation system attempts to achieve several different
  objectives: monitoring of CARERE work and performance, monitoring of PRDC
  work and performance, monitoring of the local planning process and outputs, and
  monitoring of LDF subprojects.
♦ Data in the commune databases is not updated on a regular basis. This makes
  evaluation of socio-economic impacts impossible.


   D. Reconciliation Areas

One of the most telling achievements of the SEILA program is the way it responded to
the challenge of inclusion of the former Khmer Rouge areas. Rapid trust was built up
between these former enemies, provincial staff, and CARERE management and staff, as
well as the rapidity in which the VDCs and CDCs were established.
It is also remarkable how quickly the people in the Reconciliation Areas “took” to the
participatory process of SEILA. Evidence of their enthusiasm was demonstrated by
people in the VDCs and VDCs as well as the District Governors and Deputy District
Governors. It appears to be a total success.


   E. Sustainability

The project is costly, for it not only has to provide staff for capacity building, training,
monitoring and evaluation, and management advisory services, it also has to provide
salary supplements to government staff, US $80 per month for provincial level staff and
US $40 per month to district level staff. In addition, motorcycles are provided, as staff
constantly travels to the various CDCs and VDCs. This raises great debates about the
sustainability of the project. It is perhaps the biggest issue of the project.

There are several key elements that provide a great deal of sustainability to the project:

♦ A new decentralized rural development structure is effectively providing, on the
   whole, good quality infrastructure throughout its target areas.

♦ Provincial administrations, through the PRDC and its other bodies, are continuing to
   build the capacity necessary to continue this process. National staff in both CARERE
   and government are increasingly taking its own initiatives and taking over more of
   the management of the projects. As this process continues, CARERE can reduce its
   technical support, training, and catalytic involvement, and take more of an advisory
   role.

♦ Already, two of the five Provincial Program Managers resigned, turning the positions
   over to Cambodians. Next year, the PPM in Siem Reap will step down to be replaced
   by a Cambodian.
                                                                                            x


♦ Managing rural development is increasingly being taken over by the ExCom; and
   technical support and facilitation is already provided by the Secretariat.

♦ CDCs and VDCs are demonstrating the ability to make plans for their communities in
   a participatory and transparent manner.

♦ Beneficiaries are consistently making adequate local contribution to all projects.

♦ The Royal Government of Cambodia has officially recognized the RDS structure,
   which is directed by SEILA, as the official structure in the five provinces in which
   SEILA operates.

♦ The Royal Government of Cambodia funded over $200,000 in 1999 and is
   considering $400,000 to $500,000 for the year 2,000.

♦ All local projects implemented by the various ministries in Ratanakiri will require a
   10% local contribution, and in Battambang all road projects will require a 20% local
   contribution.

If, on the other hand, the SEILA/CARERE/LDF project is simply going to be one stand-
alone project that eventually needs to be phased out, different strategies need to be
addressed. Much of this is dependent on how quickly the RGC moves on the
administration reformed and demobilization programs. There is a huge potential of funds
related to reducing the civil servants and army personnel, both of which are totally
overstaffed. In addition, the new Commune Councils will have taxing powers, and there
are considerations for taxing land in provincial and village towns.

The other issue regarding sustainability is the situation in Cambodia. Because it has not
only had its infrastructure devastated, but also the last thirty years of civil war has had a
tremendous impact on the capacity of human resources and institutions, the economy,
society, culture, religion, and the sense of community. Basic trust between people in
villages has been affected. Most of the educated people were either eliminated, died of
starvation, or left the country. The same fate happened to over 75% of the teachers. This
makes the situation in Cambodia unique from most other developing countries, and
requires much more time and effort in capacity building, training, technical support and
advisory services which is extremely time- and- people intensive and, therefore,
expensive. This is especially true when you are literally changing the way local
government operates and delivering vitally needed infrastructure at the same time.


   G. Transition Strategies
LDF funding is due to end this year while the SEILA/CARERE project is funded through
the year 2000. The project needs LDF funding to take it through the year 2000.
                                                                                           xi


Year 2000 should be spent consolidating current activities, bringing the main
experimental aspects of the local planning program to a close (perhaps offering a series of
options within the model). SEILA/CARERE/LDF should decide on how many
communes and villages they can practically bring under the LPP and LDF and complete
strategic plans with time frames and budgets that address this. It is also important that
CARERE take measures to avoid losing staff during the period of uncertainty before
another project is signed, namely ensuring that the formulation of the next phase is
completed as early in 2000 as possible.

Emphasis in the year 2000 should also focus on building training-of-trainers’ capacity
within the facilitation teams at the provincial level, on working with the PRDC to
establish how best to organize the facilitation workload and supervisory responsibilities,
and on developing a clear strategy that will allow the PFTs and DFTs to reduce their role
in mature communes.

It is quite conceivable that within three to five years, CARERE could cease to exist, as it
is known now. The awkward problem is that money is needed to fund provincial, district,
commune and village programs and projects to provide work for the government staff
who have been receiving this training and support. The institutional structures are in
place and are functioning, and the government staff has the capacity to perform the work,
as do the members of the various committees. Funds need to come from donors and
eventually from government, otherwise, the institutions that have been established will
lack one of the key elements that make their existence meaningful.

Any exit strategies within the next five years will be difficult, particularly if the
government wishes to replicate the process in other provinces. However, if the
Administration Reform and Demobilization Programs begin to be implemented, funds
from those salaries can be channeled into higher salaries for the civil servants that actual
work full time. As these funds continue to increase over time, LDF and other funds can
be proportionally reduced. In addition, as stated above, government staff is increasingly
taking over more responsibility. CARERE can continue to play more of a true consulting
role rather than a continuous trainer and support role.

This can also be true if SEILA/CARERE is called upon by government to help other
provinces replicate the project. SEILA/CARERE will play a lesser role in their own
provinces while sending key staff into other targeted provinces to assist in the
establishment of the SEILA program.

If donor funds are reduced sharply over the next few years, and little or no money comes
from government, it is likely that the VDCs will not sustain themselves. On the other
hand, with the formation of the Commune Councils as administration units of
government; it seems likely that they will be sustained. This is particularly true if they
are granted taxing powers, which is likely. Government is considering taxing real estate
in provinces, districts and even small towns. This, plus a market tax and even some form
of income tax for families with large pieces of fertile land, draft animals, etc. could
generate income for development.
                                                                                          xii




   H. Recommendations
There is a need to emphasize villages and communes doing more self- help projects
themselves, but with the technical support of local governments. It should be made clear
to CDCs and VDCs from the beginning what the government will do and where their
responsibility ends. Village responsibilities should be clearly stated. The allocation
process could be altered somewhat to reward villages and communes that initiate and
complete projects themselves. This could be done during the second and third cycles.
This would perhaps help people to build their sense of community while at the same time
developing more of a self-help attitude.

More emphasis could be put on villages seeking outside help or doing more work
themselves on projects where LDF allocations were insufficient to cover total costs. For
instance, if the allocation for a village was only $4,000, the project cost $7,000, and the
village wanted it badly enough, they should be encouraged to seek other funds or do more
of the work themselves.

From the beginning, there has been much skepticism and criticism from the NGO
community about the SEILA/CARERE/LDF project. Many of the NGOs still view the
project from the standpoint of “community development” rather than a capacity building,
decentralization and local government. There appears to be a strong need to inform the
NGO community of the real mission of the project. The ET realizes this is a difficult
situation and not easy to correct; however, it is hoped that the project would develop a
public relations strategy that will more fully address this issue.

VDCs often do not represent the majority of the people in a village. The members are
usually elected because they can perhaps read and write a little, and this is thought
necessary to submit projects to the CDC. The VDC is often not connected with the Wat
Committee or other groups that may have influence in the village. New strategies should
be developed whereby more of the total village is represented and where more inclusion
of traditional leaders is advanced.

Where NGOs have assisted in establishing small committees, there is often the complaint
that the newly established VDCs often ignore these committees. It is recommended that
part of the orientation training deal with this issue, and that a formal mechanism be set up
where these groups are informally part of the committee.

It should be made clear from the start that after the first few VDC meetings, all meetings
should be run by the VDC Chairperson and not the government facilitator. Also
chairpersons should be trained at the CDC level on how to conduct participatory
meetings and how to keep different groups involved.

Give villages a clear explanation of the CDC and its roles and responsibilities as well as
the LDF and how it works. In addition, show the villagers the commune map and explain
                                                                                         xiii


how it was created and what it means so they have a clear understanding of the whole
process.

Villagers should be notified well in advance of all VDC meetings, and should always be
involved in decisions on such things as where wells and ponds should be located, how
contributions should be collected and how projects will be maintained.

The members of the SEILA team should consider spending the year 2000 consolidating
current activities, bringing the main experimental aspects of the local planning program
to a close (perhaps offering a series of options within the model), and thinking about
ways to work strategically given existing constraints and demands. It is also important
that CARERE take measures to avoid losing staff during the period of uncertainty before
another project is signed.

The three provinces visited by the ET demonstrated that they can probably complete the
expansion of VDCs and CDC within two years. As they move to set up new committees,
the older committees become more mature and need less attention from government staff
(PFTs, DFTs and TSSs); therefore, this staff can spend more time assisting new VDCs
and CDCs. This is also true of CARERE’s local capacity building staff. One important
exception is that new members of VDCs will need training and assistance. This is also
true where, under a new election, all of the elected members are new. This suggests the
gradual reduction of both CARERE and government staff over time. It also means that as
government staff matures in proficiency, less is needed of CARERE, since they are there
primarily as trainers, support providers, and advisors.


   I. Future Project formulation

UNCDF/UNDP or other donors could provide a full time local government and
decentralization specialist for one year to work between SEILA, CARERE and national
government to assist them in the formulation of plans to take the project into 2,001 and
beyond. The role would be for a technical advisor in local government and
decentralization and would require facilitation skills. The person would live in the
country, network between different agencies, serve as a catalyst, and assist in the
development of a three- to- five- year strategic plan to which donors could respond.
Some of the possible issues the person would deal with are:

♦ The experience to date has been that the LPP process has been modified in each of the
   SEILA provinces to suit the local context. In the next phase of the project, it will be
   important to begin developing a general framework for national guidelines that still
   allows local flexibility.

♦ Rather than reduce present SEILA staff, it is possible that staff should be increased to
   provide key people to go into other provinces to train local leaders in the
   methodology.
                                                                                         xiv


♦ Reinforce the new Decentralized Development Fund at the national level that will be
     a depository from an expanded list of donors from which SEILA could work. This
     would include more strategies for making SEILA the vehicle for delivering
     development through local government that many donors could respond to.

♦ SEILA has established a decentralized planning and financial structure that has
     proven effective and is potentially a national model. It is hoped that the model will
     continue to gain support from the Royal Government and will be allowed to adjust to
     changes in the local structures anticipated by the upcoming Commune Elections, all
     the while maintaining responsibility for coordination of development activities at the
     provincial level. This would mean that the elected Commune Councils would
     coordinate their development activities through a non-elected body.

♦ With the establishment of the new Commune Councils, there is a need to identify and
     coordinate outside financial resources for development projects, including multi-
     sectoral, at the commune level.



I.    CONCLUSION

SEILA/CARERE/LDF has been an enormously ambitious project. It has also been
costly. It has attempted to almost remake a local government process and deliver vitally
needed infrastructure development at the same. Although there is scope for improvement
and much hard work yet to be done to institutionalize the structures and processes that
have been created, the creation of this system has been a major achievement and has for
the most part achieved its ambition. The SEILA/CARERE/LDF system has
accomplished the following:

1. The project has more than achieved all of its goals and objectives and the outputs
   described in the main phase document.

2. It has assisted the governments in the five SEILA provinces in establishing a
   participatory decentralized local government planning and development system in a
   government/society known for its top down hierarchical style of rule.

3. It has implemented a precedent in local government in Cambodia whereby a certain
   percentage of government committee members must be women.

4. The system created has been formally accepted by the RGC and is looked upon by
   most government officials as the model for decentralized local government in all
   provinces.

It has had a major influence in the drafting of new local government laws and with the
formulation of the Commune Councils in the year 2,000, which is seen as move towards
democratization of local government.
                                                                                       xv



The major task of SEILA/CARERE/LDF is now to consolidate progress and to assist and
work with the Royal Government of Cambodia, as the government desires, and other
donors in the formulation of strategies and plans to move the SEILA/CARERE/LDF
project forward into the year 2,000 and beyond. In addition, to assist the government, if
it so desires, in its move towards decentralization in other provinces.
UNCDF Cambodia Final Evaluation                                                             2




1.     INTRODUCTION

     1.1. Purpose of Evaluation

The overall objective of this final evaluation is to assist the Royal Government of Cambodia,
beneficiaries, UNCDF, and UNDP to:

♦ Assess the efficiency, effectiveness, relevance and impact of the project, and make
     recommendations for improvements wherever possible.

♦ Provide feedback to all parties to improve the policy, planning, project formulation, appraisal
     and implementation in the context of the next phase projections of the SEILA Program.

♦ Gather extensive lessons learned and operational recommendations that will feed into the
     future orientation and joint-formulation of the SEILA Program beyond 2000 with UNDP,
     SIDA and other possible partners. Recommendations should propose ways of ensuring a
     smooth transition of the project to the next phase and the further institutionalization of the
     project into the governmental structure taking into account the emergence of the “Commune
     Councils.”

♦ Draw lessons that will improve the efficiency, effectiveness, relevance and impact of other
     UNCDF-funded LDF projects.


     1.2. The Cambodian Context

Much has been written about the horrors of continual war and civil strife in Cambodia during the
last thirty years and their effects on the infrastructure, the capacity of human resources and
institutions, the economy, society, culture, religion, and the sense of community. Basic trust
between people in villages has even been affected These events have had an enormous influence
on the process of development as well as on the results. Some of the issues that should be
considered when implementing development in Cambodia include the following:

♦ Most of the educated people were eliminated, died of starvation or left the country.

♦ The same fate happened to over 75% of the teachers.

♦ 85% of Cambodians are subsistence farmers with little education, rarely going beyond
     primary school. Basic concepts that are usually taken for granted in other countries often
     cannot be assumed in Cambodia. As a project manager said, “The project was designed to
     last one year, and after three months we realized that it just wasn’t working. We had to
     almost start over and teach basic skills in problem solving and planning that we had assumed
     they knew. Now we know the project will need to be extended two more years.”
UNCDF Cambodia Final Evaluation                                                                      3



♦ Posttraumatic stress syndrome is still an issue for most Cambodians who are over twenty-five
      years old, and it can manifest itself anytime in people affected. This condition was caused
      not only by war but also by the fear that a neighbor or friend could turn you in to the
      authorities for termination to save their own lives.

♦ Many people are still reluctant to join or be a part of a committee, cooperative or association
      because in the past this was something one was ordered to do. It usually meant more work
      and regimentation.

♦ The same suspicion is true towards government. One hears village people say, ”When they
      come into a village, it only means that they want us to do something or take something away.
      They never give nor ask for advice. They are the outsiders.”

♦ Studies have shown and development workers have affirmed that Cambodia is an extremely
      hierarchical society where simple obedience is expected without asking questions or sharing
      ideas. Krishnamurthy states in a study of two villages, “Village meetings that foster
      collective discussions and decision-making do not exist.” - - - “Leaders assume compete
      authority and responsibility over village affairs, and their decisions are accepted without
      question though questions exist in people’s minds.”1

♦ During the years 1975-79, the roots of social order and traditions were transformed, “For
      Khmer, the DK (Democratic Kampuchea) period is viewed as a time when the core elements
      of Khmer social order were shaken to their roots. Basic elements of social status: age, sex,
      wealth and religious practice were called in question and reversed, separated from the old
      social order but not reintegrated into a new social order.” (Ledgerwood 1990:210)

♦ Instead of villages set in clusters as in most places in Southeast Asia, increasingly
      Cambodian houses are strung along a road or path in an ill-defined manner, which sometimes
      results from a lack of what westerners refer to as “community.”

♦ In regard to the above, a recent study shows that individualism, particularly among the
      poorest, is on the rise. Surviving in an economy with little food is one cause. Struggling to
      adapt to a market economy is the other.

♦ The commune, created by the French Colonial Administration, was used rather arbitrarily to
      section off areas for control purposes. Since then it has been used mainly for the Commune
      Chief to recruit men to fight skirmishes or to mobilize people for work crews. There is little
      or no indication that the villages are connected to a larger community.

Compared with other countries, experienced development workers in Cambodia are adamant in
advising that projects often take much longer and outputs are sometimes less satisfactory.
However, they often also say that by taking a longer view on projects and by building capacities
at the beginning, the resulting outcome can be more than adequate.


1
    Krishnamurthy, V. “The Impact of Armed Conflict on Social Capital”, World Bank Study, March 99
UNCDF Cambodia Final Evaluation                                                             4


   1.3. Background


       1.3.1. UNCDF and CARERE 2

Following the Paris Peace Accord of 1991, hundreds of thousands of refugees moved into the
province of Battambang and then into Banteay Meanchey, Seam Reap and Pursat in the
northwest part of Cambodia. In 1992, the first phase of CARERE (Cambodian Area
Resettlement and Reintegration Project) was established by UNDP in the provinces of
Battambang and Banteay Meanchey. Its focus was on emergency relief and quick impact
projects directed primarily at the refugees and the villages in which they were placed.

As Phase One was ending, there was a general feeling that CARERE 2 (the second phase
hereinafter referred to as CARERE) should switch from emergency relief to a more
developmental mode. In pursuit of this, a March-April 1995 UNCDF mission in Cambodia
recommended the establishment of a Local Development Fund (LDF) to be used with the Local
Planning Process (LPP) in a pilot project in two communes in each of two provinces,
Battambang and Banteay Meanchey. This was to be used as a policy experiment. Later the
decision to expand into the provinces of Pursat and Siem Reap in the northwest and Ratanakiri in
the northeast was made in September 1998. The overall objective was to assist the Government
of Cambodia as it develops critically needed rural infrastructure to support local economic
development through a decentralized, participatory decision-making and implementation
process. The LDF would provide funds for rural infrastructure projects while the LPP would be
the planning vehicle that would support the process. The provincial government staff,
principally the provincial departments of rural development (PDRD), would be the
implementors, while CARERE would supply most of the training and technical support and
provide catalytic force to move the project along.

The Main Phase Project, Document No. CMB/97/CO1, stated the Development Objective as, To
alleviate poverty and spread peace in Cambodia by strengthening the bonds linking society to
the structures of the State and empowering the Cambodian rural population to become fully
participating members in the development process through decentralized governance.

It also provided for five immediate objectives:

1. To develop a participatory, demand-driven process of rural development, infrastructure
   planning and implementation.

2. To fund the construction of priority rural infrastructure projects identified through this
   participatory process.

3. To augment and institutionalize the capacity of the PRDCs.

4. To augment and institutionalize the capacity of the VDCs and CDCs.
UNCDF Cambodia Final Evaluation                                                               5


5. To establish incentives for increased internal support for rural infrastructure projects through
   central government financial, technical and in-kind contributions and from local
   communities, while at the same time developing a sense of “ownership” of the local projects.

The funding for the pilot phase was approved by UNCDF in November 1995, and training began
in March 1996. Implementation began shortly after the training had finished with the first set of
projects completed in July and August. During the course of the pilot phase, several technical-
assistance and monitoring missions were conducted by UNCDF with the cooperation of
CARERE.

UNCDF provided a budget of 4.526 million dollars, 3 million of which would go to the LDF for
infrastructure development. UNDP would provide 20.5 million primarily for the operation of
CARERE.

This was the beginning of a policy experiment to develop a “model” for decentralized rural
development that could perhaps be replicated in other provinces. It therefore had the following
features:

         •   Financial resources, planning and implementation of development from central
             government to local government.

         •   Participatory local government institutions that would enhance this devolution
             process.

         •    Management, planning, implementation, monitoring and evaluation capacities of
             these institutions and the staff within.

         •   A transparent and workable financial management system.

         •   A process for delivering sustainable infrastructure and other rural development
             projects.


         1.3.2. The Emergence of SEILA/CARERE/LDF

At the national level of government, the SEILA Task Force was created by the Council of
Ministers to oversee the CARERE/LDF project. The original five ministries included on the task
force were the Ministries of Interior, Rural Development, Planning, Finance, and Women’s
Affairs. In 1999, the original CARERE 2 document2 was amended to include the Ministry of
Agriculture, which extended the project through the year 2000.

The amended CARERE document also stressed that the process needs to move away from the
implementation modality of a UNDP project to one managed by government using its own
systems, with CARERE assisting in the design of systems and providing adequate support,
safeguards and transparency.

2
    UNDP numbers CMB/95/011/L/01/31 and CMB 95/A11/A/XX/31
UNCDF Cambodia Final Evaluation                                                             6




A Sub-Decree in December of 1997 from the Royal Government of Cambodia officially
recognized the SEILA Task Force (STF) and the institutions and processes under which it
worked. This assured the authenticity of the project.

The mission of the Evaluation Team (ET) is to evaluate the LDF Project. As the ET proceeded
through its work, it became apparent how explicitly the project was tied to the performance of
SEILA/CARERE and of the local governments involved. It also became apparent that the
project had major policy implications at the national level regarding the future of local
government. At this time there is much support at national government level to replicate the
SEILA/CARERE/LDF model in other provinces. In addition, laws are being drafted by the
Royal Government of Cambodia to establish new units of local government that will impact the
project. This will be discussed later in the report under Commune Councils.



    1.4. Composition of Mission

The Evaluating Team is comprised of the following:

•   Robert Leonard, team leader, (a local government and institutional consultant who lived in
    Cambodia for almost four years), has trained all of the provincial Governors and First Deputy
    Governors and all of the top leaders in each of eleven provincial offices in management and
    planning.

•   Dr. Robert Guild, an infrastructure specialist and engineer who recently completed an
    evaluation of the Vietnam Rural Infrastructure Development Fund for UNCDF.

•   Cristina Mansfield, a community participation consultant with four year’s experience in
    Cambodia working with international and local non-governmental organizations (NGOs).

    1.5. Terms of Reference

The Evaluation Team has discussed the detailed Terms of Reference shown in the Annex I with
several staff members of UNCDF, and a general consensus has emerged regarding the major
thrusts or focus for the evaluation and recommendations. Below is a summary of the team’s
understanding of that general consensus:

♦ Given the number of evaluations and findings already made on this project, and considering
    the need to provide input into the thinking for a new formulation in early 2000, the primary
    aim was to provide a comprehensive view of lessons learned with particular attention to any
    new lessons not yet revealed.

♦ The major emphasis of the report was on providing recommendations as to how to apply
    those lessons learned to strengthen the project as it now stands, and more importantly, to
UNCDF Cambodia Final Evaluation                                                               7


   show how they can be applied to the development of the conceptualization of both the new
   Commune and Administration Law being drafted and the upcoming project formulation.

♦ In addition, from the lessons learned, the focus was on how to replicate the project
   experience to other provinces and in the local government/decentralization framework.

♦ Infrastructure and other projects was examined for their quality of construction,
   appropriateness, impact, sustainability and effective use of resources. Particular attention
   will be given to local participation and contributions, sense of ownership, the decision-
   making process, and project identification, appraisal and other related matters. In addition,
   attention will be given to the appropriateness of the projects and their impacts.

♦ A close examination was made of the flow of funds, financial management, transparency,
   and procurement procedures.

♦ To assess the viability of the institutions that have been established, the team reviewed such
   things as participation (including gender and the poorest of the poor), sense of ownership,
   openness, transparency, problem solving and decision-making processes and involvement in
   planning, implementation, monitoring and evaluation.

♦ As part of the examination of the planning process and level of capacity, the team reviewed
   the suitability and accessibility of training manuals and procedures and how much has been
   absorbed by individuals at all levels.

♦ The Evaluation Team examined the transfer of skills from the project to the government
   structures and the possibility of exit strategies.

One of the most important aspects of the evaluation is the strategic analysis. The main issues
are:

       •   The sustainability of both the decentralized planning structure and of local and
           national finances.

       •   The potential for extension of the LPP and the decentralized rural planning structure
           to serve as a national model and for extension to other provinces.

       •   The potential for the LPP and the decentralized rural planning structure to serve as a
           model for other donors and for activities beyond rural infrastructure.

       •   Risks and opportunities for adapting the LPP and the decentralized rural planning
           structure to the proposed new commune councils.


   1.6. Methodology

The methodology of the evaluation comprised four basic activities:
UNCDF Cambodia Final Evaluation                                                          8




♦ Discussions with CARERE senior management and PSO staff, PRDC Executive
  Committees, and units of the Secretariat (Facilitation, Finance, and Technical Support) to
  understand the program, roles, challenges, opportunities, and suggestions for the future, as
  well as to assess their capacity and views about moving to scale.

♦ On- field visits met with district, commune, and village development committees. Village
  and commune leaders and villagers attended district integration workshops and bidding
  meetings and inspected project sites.

♦ Reviewed the extensive project documentation, including previous evaluations, analysis of
  the commune and LDF sub-project databases.

♦ Meetings with NGOs in the field and in provincial towns, and assessed past collaboration and
  identified potential areas for future collaboration.

♦ Debriefing meetings with PRDC Executive Committees, provincial staff, PSO staff and
  CARERE staff.
UNCDF Cambodia Final Evaluation                                                            9




2.     DESCRIPTION OF MAIN PHASE LDF PROJECT

The information set forth below is extracted from the UNCDF Project Document No.
CMB/97/CO1 (Project Document).

The project is a continuation of the pilot project in two provinces for the establishment of a Local
Development Fund with an initial allocation by a grant from UNCDF of US $3 million over a
period of three years from 1997 - 1999. Given the significant and positive lessons learned during
the pilot phase and the adoption of the LDF/LPP methodology by CARERE, this amount is
increased by US $1,426,400. This is to allow for a greater amount of infrastructure projects and
to provide for two Technical Assistance Advisors and a Technical Backstopping Monitoring
arrangement with a nearby technical institution.

With two communes in each of the two provinces already receiving LDF funding, the project
provides for an expansion into 8 – 10 additional communes in 1997, and at least that in 1998 and
1999. The grants support the process of decentralized planning under the LPP in the form of
Indicative Planning Figures (IPFs) to the CDCs in the amount US $50,000 for projects that have
characteristics of “public goods.” The funds are allocated over a three-year cycle. To qualify,
CDCs must have complete coverage VDCs or similar village committees sponsored by NGOs.
The specific allocation to each commune depends on the population size, the number of villages,
and the amount of development that has already taken place. Communes and villages receiving
projects are to make at least a 10 % contribution.


     2.1. Rationale

The interrelated rationales for undertaking the main phase of the Cambodian LDF project are the
following:

1. Infrastructure development is well known to play an important role in poverty alleviation and
   economic development. The project will result in the provision of critical types of
   infrastructure in the areas of coverage.

2. During the past few years, the Government of Cambodia has placed an increasing emphasis
   on furthering rural development.

3. The government has embarked on a major program of decentralization and participation in
   the development decision-making process.

4. Government resources to implement its intended programs of infrastructure provision, rural
   development, and decentralized planning and implementation of projects are extremely
   limited.

5. The establishment of the LDF is a critical component of UNCDF/UNDP efforts to reorient
   their broader program of activities in Cambodia.
UNCDF Cambodia Final Evaluation                                                            10




In summary, the core rationale for the project is to assist the Government for Cambodia as it
develops critically needed rural infrastructure to support local economic development through a
decentralized, participatory decision-making and implementation process. Much of the rationale
is based on the growing confidence of Cambodian counterparts in implementing the LPP, which
they found to be workable and fully replicable for the main phase.


   2.2. Development Objective

To alleviate poverty and spread peace in Cambodia by strengthening the bonds linking society to
the structures of the State and empowering the Cambodian rural population to become fully
participating members in the development process through decentralized governance.


   2.3. Immediate Objectives

The project is centered on the five immediate objectives:

1. To develop a participatory, demand-driven process of rural development, infrastructure
   planning and implementation.

2. To fund the construction of priority rural infrastructure projects identified through this
   participatory process.

3. To augment and institutionalize the capacity of the PRDCs.

4. To augment and institutionalize the capacity of the VDCs and CDCs.

5. To establish incentives for increased internal support for rural infrastructure projects through
   financial and technical contributions of central government financial and in-kind
   contributions and from local communities, while at the same time developing a sense of
   “ownership” of the local projects.


   2.4. Outputs

The major outputs of the LDF project include the following:

1. The establishment of a pilot to test and further develop proposed LDF procedures in two
   provinces of Battambang and Banteay Meanchey (pilot phase), now expanded to include
   another three provinces.

2. The establishment and institutionalization of the LDF in a suitable agency at the provincial
   level following an evaluation of the pilot phase.
UNCDF Cambodia Final Evaluation                                                           11


3. The development of appropriate procedures and proposals for developing rural infrastructure,
   including: a program of training for PRDC, DRDC, CDC and VDC members that supports
   the institutional arrangements and procedures of the LDF outlined above; and
   institutionalization of the intergovernmental project planning and implementation process
   outlined above.

4. Physical outputs created by means of the LDF process.



   2.5. Activities, Project Inputs and Implementation Arrangements

To assist CDCs and VDCs who have entered the LPP system, Local Planning and Technical
Support Staff drawn from the provincial and district line agencies will train and support the
effort. These LPS and TSS teams will be expanded to 5-6 teams of 3-4 persons each in the
initiation of the planning process. These teams will be under the direction of the Secretariat and
will be provided a set of motorcycles. After the experience in the pilot phase, the LPP was
shortened to the following six major steps:

1. General Orientation – The LPS team will conduct a two- day workshop at the CDC followed
   by a three -day workshop at the village level explaining the details of the LPP and the roles
   of the VDCs, CDCs, DDCs and the PRDC in the LPP.

2. Data collection and analysis for formulation of Village Development Plan (VDP) – Using the
   Rapid Rural Appraisal (RRA) and Participatory Rural Appraisal (PRA) approaches, the
   VDP, including a village map, will be developed over a five-day period.

3. CDC aggregation of VDPs into Commune Development Plan (CDP) – VDPs with data
   collected at village level will be brought together at the CDC where the CDP will be
   developed over a four-day period.

4. Project formulation – Proposals will be prepared for the project(s), including a basic
   feasibility study and cost estimate. A portion of the IPF (up to 15 %) could be used to defray
   the costs of planning and project preparation. Up to 5 % could go to the CDC and VDC as
   administrative fees and up to 10 % to the TSS, sectoral ministries, NGOs or private firms.

5. Project appraisal – CDC will submit selected proposals through the DDC and then to the
   PRDC for approval.

6. Approval – The PRDC will return the proposals to the CDCs through the CDCs indicating
   which proposals were approved and what action should be taken on other proposals.

After receiving approval from the PRDC, the CDC/VDCs will organize tasks and activities for
project implementation as well as monitoring and maintenance. On larger projects requiring
more people and technical support, two committees will be set up. The technical committee will
work with the contractor while the other will mobilize workers and organize logistics. The need
UNCDF Cambodia Final Evaluation                                                              12


for technical assistance in project design is recognized to vary considerably by type of project,
and it is intended that project “templates” will be developed setting forth technical and cost
standards.

Training of the LPS/TSS teams will be for ten days and will include such things as review of the
LPP; RRA and PRA techniques, project feasibility study, project formulation, implementation
and management, financial and contractual procedures, and monitoring and evaluation. At the
start of the implementation of the LPP, a series of information and training workshops will be
given to the provincial and district staff to foster commitment of the government in the process.

Disbursement activities will follow the Ministry of Finance (MoF) guidelines and will be subject
to CARERE monitoring.


   2.6. Monitoring and Evaluation

CARERE Local Capacity Building Staff (LCB), supported by the LDF M & E Advisor and the
CARERE M& E staff in Phnom Penh, will assist in the monitoring of projects as well as the
progress of the LPP. Because of the risky and experimental nature of the LDF project in
Cambodia, there should be a special, more intensive prototype M & E system designed using the
learning experiences in the pilot project and in the ongoing process of the LPP. CDF and
CARERE/SEILA will use the same M & E system. The system will, under the guidance of the
CARERE M & E Advisor, produce an M & E work-plan with performance indicators. Baseline
data will be collected on the basis of the indicators and will show the source, baseline
information on each indicator, periodicity and duration of collection, who will collect it and cost.
A Stakeholder Informant Group will be identified at the inception of the project consisting of
individuals or groups who are affected by the project. In addition, focus group discussions with
villagers concerning the outcome of selected micro-projects will be conducted as well as case
studies by third parties. Studies will be carried out by external consultants, who will train
relevant CARERE staff, who in turn will train their counterparts. In addition, staff involved in
the LPP will keep a journal of observations during the planning rounds. As the project moves
into the main phase, the burden of capacity building will shift from short-term consultants to the
CARERE provincial staff.


   2.7. Risks

The risks associated with the project along with the estimated degree are as listed in the table
below:


                                       RISKS                                        PERCEIVED
                                                                                     DEGREE
      1. Many weaknesses in the governmental system, lack of legal and
      institutional framework for efficient government, and no recent history of   Medium to high
      local government.
      2. Because of the lack of RGC funds for public investment, many types of
      funding programs are being delivered by international organizations with
      different types of planning approaches and financial procedures.
      Institutionalizing a consistent and uniform planning system over time is a
      great challenge.
UNCDF Cambodia Final Evaluation                                                               13


      funding programs are being delivered by international organizations with
      different types of planning approaches and financial procedures.              Medium
      Institutionalizing a consistent and uniform planning system over time is a
      great challenge.
      3. Cambodia is desperately short of human, financial and institutional
      resources, and the central government is not in a position to allocate        Medium to high
      resources for adequate operation and maintenance of infrastructure.
      4. Because people have become dependent on external agencies for
      survival and have been traumatized for many years, there is a lack of self-   Medium to high
      reliance.
      5. Several areas of the northwestern provinces are still insecure related
      to opposing groups, and land mines hinder infrastructure and agricultural     Medium to high
      development.

      6. There is no guarantee that the participatory decision-making process
      will bring consensus or transparency, and subsequent rounds of the LPP        Medium
      may have adverse effects on financial contributions and community
      participation.
      7. Training and capacity building is dependent on the success of the
      CARERE staff to train and re-train LPP and TSS in the PRDC. There is a        Medium
      need to seek commitment from central and provincial authorities that
      transfers will be kept to a minimum.
      8. There may be too many projects for the capacity of the planning and
      technical staff of the PRDC.                                                  Low
      9. Migration of staff of the PRDC to other donor agencies.                    Low
      CDF could lose its identity and “control” under such a large UNDP project.    Low




   2.8. Changes in Project Implementation Since Approval

SEILA/CARERE/LDF demonstrated that with proper funds for infrastructure development,
capacity building and technical support, both lower as well as provincial levels of government
could manage a wide range of development activities. In addition, this process effectively
supported a wide range of newly created local government development institutions. This
convinced national government that this decentralization process was an effective approach to
local development. To some degree, this has led the RGC to charge the MoI with drafting of
both the Commune Election Law and the Commune Administration Law.

Given UNCDF’s mandate for support to decentralization, as well as the strategic role the agency
has played since the outset of the SEILA Program, UNCDF made the decision to support the
development of a national policy on decentralized development in September 1998. A mission
was launched in December 1998 to formulate a project for Assistance to the MoI to Draft the
Commune Council’s Basic Legislation. By February 1999, it was agreed that the activities would
be funded from CMB/97/C01; and a South African lawyer, the author of the formulation report,
returned to Cambodia to support MoI’s inter-ministerial workshop on decentralization and
commune councils. He is providing ongoing support to MoI for the drafting of both laws.
Additional support on the policy underlying the Commune Administration Law has been
provided through short-term missions taken by AIT, with a local finance expert from MIT, and
UNCDF’s Chief Technical Advisor for the LDF.
UNCDF Cambodia Final Evaluation                                                        14




With activities proceeding apace at the national level in creating a policy framework for
decentralized development, the need became immediate for SEILA/CARERE/LDF to accelerate
its work in perfecting and testing sustainable and replicable development planning and
implementation systems at the local level. To support this effort, UNCDF agreed both to expand
the coverage of the LDF to the remaining three SEILA/CARERE provinces (Pursat, Rattanakiri
and Siem Reap) and to extend the contract of the Senior M&E Advisor through the end of
October 1999. All of these decisions were based on recommendations made in the End of
Mission Memorandum, which was produced following the UNCDF-AIT Mission (20 August – 3
September 1998) for Review of the Second Cycle of the Local Planning Process. No additional
capital investment funds were required to provide the additional coverage, but a budget revision
(BR “E”) was prepared to cover the costs of the M&E expert.

Rather than completing the LDF at the end of 1999, UNCDF agreed to carry the project through
a transition phase from January through December 2000 to bring the project cycle in line with
that of the UNDP project. In discussions held at the CARERE Tri-partite Review of 20 May
1999, UNCDF clarified that the purpose of this phase would be to maintain a minimal strategic
presence and to provide support to key areas until a joint formulation (among UNDP and other
donors) for the next phase of SEILA (post-2000) could be finalized. The main focus of UNCDF
support during this transition period will be to provide minimal support to the following
activities:
§   consolidation of the M&E strategy in view of transferring the system to the government;
§   support to the government in the further development of the upstream policy dialogue and the
    preparation of the commune administration law; and
§   provision of the minimum required capital investment resources.

Consistent with these commitments, UNCDF agreed at the Tri-partite Review to continue to
support the Senior M&E Advisor through the transition period. A three-month extension of her
contract, through December 1999, has already been finalized. In addition, the contract was not
extended for the Infrastructure Advisor.

Decision- making regarding the other areas of UNCDF interest has continued since May.
CARERE submitted to UNCDF a proposal for capital investment for the year 2000 in June 1999.
At UNCDF’s request, however, CARERE has been revising the proposal, as support has become
available from the government and other donors. Discussions in support of additional technical
assistance needs have been communicated to the mission of UNCDF’s Chief Technical Advisor
in August 1999 and his subsequent Back-to-Office Report. During the September Week-in-
Residence, UNCDF brought to headquarters the legal and local finance advisors to discuss with
senior management, the Program Manager and Chief Technical Advisor the additional
anticipated needs.
UNCDF Cambodia Final Evaluation                                                              15




3.    INSTITUTIONS AND CAPACITY BUILDING – FINDINGS


        3.1.1. Provincial Governments Pre- UNTAC

Before UNTAC (United Nations Transitional Authority in Cambodia), the Cambodian
government had some aspects of strong decentralization; namely, the various provinces were
considered almost independent kingdoms or “fiefdoms” with power centered in the Provincial
Governors. Revenues were collected at the provincial level and were “sent up” to central
government.

This situation was reversed under the new government of Cambodia after the 1993 election
under the tutelage of Sam Rainsy as Minister of Finance. The primary purpose of this was to put
all government revenues into the hands of the Ministry of Finance MoF) where funds could be
controlled, the Riel stabilized, and transparency established. Under the new system, all money
was collected at the central level and then allocated to the various central ministries who in turn
developed programs and projects to be implemented in the provinces. The projects were
developed by the various ministries at the central government and then sent down to their
respective subordinates at the provincial level to be implemented. There was always a question
as to who had the final say on projects under this arrangement: the central line ministry, e.g.,
education, agriculture or health, or the governor. This was further confused because the Ministry
of Planning, in the center of government, was supposed to control all planning. This ministry,
once under the old command- style government in the 80s, was very powerful. Now its
influence has waned under the present government.

There is little question; however, that the authority of the governor and provincial administration
has been greatly reduced regarding how and where development projects will be implemented.
In addition, the source of operating and development funds is now completely dependent on the
central government. Thus, we have seen a reversal of power from the provincial administration
to power and control in central government.

The SEILA/CARERE/LDF project became the major factor in the process of reversing this
change.


     3.2. The Provincial Governments

To fully appreciate the institutional and capacity building outputs of the project, it will be helpful
to briefly review the situation in the provinces as it existed in 1995 before the
SEILA/CARERE/LDF project. By comparing the before and after situations, one can obtain a
better understanding of the new institutions that have been created, their function, and the
capacities that have been developed to manage these institutions.

In 1996, one of the Evaluation Team’s members trained all of the governors and deputy
governors in separate two-week management and planning workshops. Following these, he
UNCDF Cambodia Final Evaluation                                                           16


completed similar trainings in each of eleven provincial offices. Among the people trained in the
provinces were the remaining top leaders in the provincial offices including the Second and
Third Deputy Governors, heads of each of the line ministries and all of the District Chiefs. The
official organigram used at that time by all provincial offices had all line ministries reporting to
one deputy governor and all District Chiefs reporting to another Deputy Governor. There was no
body set up to coordinate development such as the PRDC. There was also some question as to
whether the line ministries were responsible to the governor or directly to their respective
ministries in Phnom Penh.

When the trainer asked the governors and later the provincial officials to explain their mission or
their roles and responsibilities, they would usually bring out a long sub-decree from central
government, which contained a list of duties i.e., protecting borders, registering births and
deaths, law and order and other such activities. There was little evidence of any well thought out
development plan or any system of development in provincial offices. Rather, tentative
provincial plans came down from the Ministry of Planning regarding such things as how many
tons of rice were to be produced, how many culverts were to be repaired or the number of
irrigation canals to be dug. The Provincial Governors would comment on these plans in the way
of their constraints in implementation and other items they thought important. These would then
go back to the Ministry of Planning where they were revised and sent down to the provinces as
“The Provincial Plans.” The National Assembly approved these plans.

In addition, there was little donor money going directly to provincial governments for
development; but rather it was allocated through NGOs and, to some extent, through ministries at
central government. The Evaluation Team has no figures on this, but it is believed that little, if
any, money came to provincial governments for development. There was development, but it is
questionable how much, if any, control of the planning and implementation was done by and
through the provincial office.

In the typical provincial office, many, if not a majority of staff, would come into the office with
little or no work to perform, and many (upon arriving at the office to check in) would go directly
to other jobs to earn their living. There was scarcely money to fund projects or programs, far too
many employees for the amount of work and pitifully inadequate salaries for staff. The average
monthly salary of a government civil servant was (and still is) about US $15, whereas, it would
costs about $90 to $100 to feed and clothe a family in a province. Even today, the Director of
Planning, who is the secretary of the Executive Committee and one of the key people in the
provincial administration, makes only $20 per month. The governor makes $30. One Evaluation
Team member was told by a senior provincial official that they had about 6,000 civil servants on
the payroll in 1995 of which about 10% actually performed daily work.

In taking over a new position as head of a province and after walking through the provincial
office and seeing so little activity and so few staff, the new governor was heard to quote an old
Cambodian proverb, “The wind just blew through the doors, halls and offices.”

It is in this context that the SEILA project was launched.
UNCDF Cambodia Final Evaluation                                                              17


    3.3. The New Structure for Rural Development

Before the SEILA/CARERE/LDF project was launched, the Ministry of Rural Development
(MRD) through its Provincial Departments of Rural Development (PDRD) had already set up a
Rural Development Structure (RDS) under which the project would work. The structure,
however, was mostly in its formulation stages and had little funds to put the system to use. It did
provide local government development bodies in which SEILA/CARERE/LDF and the LPP
could implement the project. The RDS was a reversal of the RGC policies for more government
centralization described above, for it provided for democratically elected bodies at the village
level, and to some degree, representative bodies at the commune level. The main bodies in the
new RDS are the following:

♦ Village Development Committee (VDC) – is recognized by the Royal Government as an
    autonomous committee that works to ensure coordination between the Royal Government
    and civil society. The establishment of VDCs and CDCs falls under the mandate of the
    Provincial Department of Rural Development (PDRD). It is composed of from 5 to 7
    (depending on size of the population) popularly elected members of which 40 % must be
    women. Voting is done by secret ballot, and members are elected for a term of three years.
    The person receiving the most votes becomes the VDC Chair who facilitates the selection of
    members to play specific roles on the committee, such as Deputy Chief, Finance Officer,
    Information Officer and Gender Officer. The VDCs are recognized by the national
    government as Autonomous Development Bodies.

♦ Commune Develop Committee (CDC) – is composed of the VDC Chairpersons in addition to
    Deputy Commune Chief, representative of district technical offices, elected women
    representatives, a commune assistant and other local leaders such as village chiefs, who are
    invited as advisors. The committee is chaired by the government-appointed Commune Chief.
    Under the newly created Commune Councils in the year 2,000, the Chair will be elected by
    popular vote.

•   District Development Committee (DDC) – is chaired by the District Chief and is composed
    of the Chairpersons of the CDCs in the district and the heads of provincial line ministries of
    Rural Development, Agriculture, Health, Education, Planning, Social Affairs and Women
    Affairs. It also includes the heads of other district offices, the District Police Inspector and
    the District Facilitator. Its main duty in the SEILA program is to support technical services
    from the provincial line ministries and departments to the communes and villages. It is also
    responsible for organizing the District Integration Workshops.

♦ Provincial Rural Development Committee (PRDC) – is at the top of the RDS and is the
  planning and decision- making body for all development in the province. It is composed of
  people from all line ministries, District’s Chiefs and other provincial departments. It
  produces a two- year Provincial Development Plan that contains all the plans of the
  provincial line ministries.

♦ Executive Committee (Ex-Com) - was created as the executive arm of the PRDC. It includes
  the heads of the provincial line ministries of Planning, Rural Development, Finance,
UNCDF Cambodia Final Evaluation                                                        18


   Agriculture and Women Affairs and meets once or twice a week. This provides an
   instrument for bringing key people together into a decision- making and implementing body.

♦ Secretariat - a body that gives technical support to the PDRC. All project plans from the
  CDCs come to the Secretariat for review and then are sent to the Ex-Com for approval. The
  staff of the Secretariat includes the Technical Support Staff (TSS) who gives technical
  support and advice to projects and to the VDCs and CDCs. There are also provincial and
  district facilitators (PFT) and (DFT) who assist setting up VDCs and CDCs and often act as
  facilitators in their meetings. The average number of staff in a Secretariat is about sixty to
  seventy and is seconded from provincial line ministries, mostly from PDRD. The largest
  numbers of the Secretariat are facilitators. On the following page Figure 1 shows a diagram
  of the Provincial Rural Structure.
UNCDF Cambodia Final Evaluation                                                          19




                             Figure 1: Rural Development Structure




                                         GOVERNOR



                                             PRDC


                                                                        SECRETARIAT
                                             ExCom


                                              DDCs


                                              VDCs


                                              VDCs




   There is a new model, which as of this date has not been implemented. It splits the
   Secretariat back into three sectors, Rural Development, Planning, and Finance. They will
   report directly to the Ex-Com and should be a good adjustment to the somewhat overly large
   Secretariat. Rural Development will have both the Facilitation Team and Technical Services,
   while Planning will contain the M & E unit. In the old pre-SEILA system, there is no
   evidence that there were any general technical support staff (they existed only in the sectors)
   and no facilitators to work at any level of Local Government. The new model would appear
   as follows:
UNCDF Cambodia Final Evaluation                                                            20




                                       Governor



                                   PRDC                Secretary or Clerk



                                               Ex Comm




               PDRD                            PDoF                          PDoP
           Local capacity                   Finance Unit                    M&E&
            building unit                                              Information Unit,
                                                                            Admin




                                  Figure 2: New Model of Secretariat

In spite of the fact that they are not official administrative bodies, the establishment of the VDCs
and CDCs is the most far reaching of all the new structures. It has many implications, such as
participatory planning, monitoring and evaluation, problem solving, project identification,
community building, accountability and transparency, to name a few. The old system was
almost totally top down with appointed officials making the decisions with little or no
participation.


   3.4. Salary Supplements

Because of the low level of government salaries mentioned in the previous section, salary
supplements are given to the Secretariat staff in the SEILA/CARERE/LDF project. All staff
working at the provincial level are given an additional $80 per month, and those working at the
district level are given an additional $40 per month. Salary supplements have been a
controversial subject for many years throughout the international NGO community in Cambodia,
and many have attempted to reduce or eliminate them but with little success. The fact still
remains that it is hardly realistic to expect government staff to work full time for $15-$20 per
month.
UNCDF Cambodia Final Evaluation                                                             21


   3.5. Planning at Provincial Level

A consolidated rolling two year Provincial Development Plan is formulated by the PRDC that
includes formalized log frame plans (OOPP logical framework system) including budgets from
all provincial line ministries. At the time of CARERE 2’s beginning, the Evaluation Team is not
aware of any log frame planning in any of the other provinces. Rather than being too
complicated and unsuitable for provincial governments, as some have suggested, the team
considered the process highly productive. It creates a solid foundation that merges project and
financial planning together into a logical and dynamic planning system.

Two Directors of Planning and two heads of line ministries along with sector staff spoke
enthusiastically about the use of this method. It was difficult at first, but after using it for over
one year, they indicated that the ease in implementing, monitoring, and evaluation of projects
was or is much more professional. The director of one of the provincial line ministries said,
“For the first time, we know where we are going, how to get there, and through the indicators in
the log frame, how to monitor and evaluate projects.”


   3.6. Assessment


       3.6.1. Communication

One could argue that the most important institutional factor in this project has been the enormous
increase in communication from the villages through every level of government up to the PRDC.
People all through the system have been taught to brainstorm, to ask questions, and to state
opinions. It was clear on visiting VDC and CDC meetings, as well as Integration Workshops,
that people (often the most vocal were women) were speaking up, sharing ideas and problem
solving, something rarely seen in Cambodia, particularly in government bodies. This mode of
communication was evident everywhere. Higher government sectors are responding to local
government through Integration Workshops. Provincial and District technical, facilitation, and
capacity building staff are working with the VDCs and CDCs. In addition, communication has
been greatly enhanced between sectors, such as agriculture, rural development, education and
health, because of the need to develop integrated plans at the commune and village level.

It was disappointing to note that most of the high level provincial meetings were still using the
very long and narrow conference table with the Governor or District Governor, as in the case in
Malai, sitting at the head of the table. The next top officials sit directly on the sides next to him
and then the other staff, according to rank, fill out the rest of the seats in descending order. The
low ranking members, sitting at the end of the table, or even in chairs at the end of the room, not
only have trouble hearing what is being discussed, but would certainly dare not share an idea
from his/her vantage point. Communication was thus severely hampered as it has always been in
recent Cambodian government. This is also true in most ministries, even the Council of
Ministries, in Cambodia. Rather than hearing lively discussions at meetings in Cambodian
government, one usually hears speeches only from a few people at the extreme high- ranking end
of the table. This is quite difficult to change, however, as the conference rooms are designed in a
UNCDF Cambodia Final Evaluation                                                              22


long narrow fashion, and long narrow tables are usually the only conference tables available.
This hierarchical top-down approach is also part of Cambodian culture, as described above in the
Cambodian Context.


       3.6.2. Roles and Responsibilities

Government staff has a much clearer understanding of their roles and responsibilities. This is
brought about by some of the following:

•   Training in their job functions;
•   Plans to be implemented have goals, objectives, indicators, and description of activities that
    have to be followed;
•   Performance evaluations regularly given as part of the monitoring system help keep the staff
    clear on their roles and responsibilities;
•   Although the M & E system has had many problems, there has been some monitoring of staff
    functions.

Not only is there more clarity, but this process produces concrete work to be done that requires
performance monitoring. This has always been a major problem as provincial governments have
rarely been clear of what their roles and responsibilities are. This has been exacerbated by the
lack of real work.


       3.6.3. Motivation

Provincial staff is motivated by many factors already described. In addition, staff have action
plans to implement. They are more focused. Most important of all is the feeling of “ownership”
demonstrated by people throughout the system. Even though there is concern that communities
have felt that projects were from “CARERE,” rather than government, little of this appeared with
provincial government staff. In all of the meetings that the TEAM attended, CARERE officials
were never in a command role, but rather, seated themselves away from the forefront more as
observers. There appeared to be this “take charge” approach by government staff whether it was
the Governor conducting an Ex-Com meeting or a facilitator leading a session on needs-
prioritization in a village.

Salary supplements are still a huge issue, and there is little question that in spite of the high
motivation of government staff, supplements are still needed which makes the project less
sustainable than one would desire. Hertzog in his “Motivation To Work,” pointed out, In
motivation of workers many things have a higher ranking than money, but there are certain
hygiene factors that come into play, the main one being people have to be paid close to an
equivalent of what others are making in similar fields. In addition, it must be enough to
maintain his family in food, clothing and basic needs.
UNCDF Cambodia Final Evaluation                                                           23


          3.6.4. A Transparent Financial System

The establishment of a transparent financial system has set the whole process on a sound
financial footing. The new system has not been fully absorbed but is the beginning of solid
financial management in provincial government that provides transparency, good tracking and
accountability.

          3.6.5. VDCs and CDCs as Institutions

Although it is not an official administrative body of government, the CDC appears to have
established for the first time in modern Cambodian history a Commune body that to some degree
represents the villages within its area even though its focus is only on development. The CDCs
hold meetings regularly, , meet with villagers and sometimes visit villages. They approve
payments to contractors and, to some extent, monitor, evaluate, plan and review projects from
villages. Members have been trained in problem solving, project planning, roles and
responsibilities, governance, human rights and a host of other subjects. It is unfortunate that the
Commune Chief is also the CDC Chief, as this has often led to his/her dominance in
participation and in selection of projects. As stated before, this will change after the CC
elections in the year 2,000. The lack of capacity is still a major problem, as most of these roles
and responsibilities are almost totally new; however, given the lack of any previous institutional
experience both by government and the members of CDCs, much progress has been made.

Like the CDC, the establishment of the VDCs as the participatory local government planning
vehicle is a bold and unprecedented move in Cambodia, bringing people together in a way rarely
experienced there. People are freely sharing ideas in meetings; women are participating; people
are learning mapping and rudiments of needs assessment, project formulation and planning.
There are, however, many issues that have arisen regarding the VDCs, most of which will be
addressed under the Local Planning Process section of this report. There are some overall
institutional issues that will be mentioned here.

As the ITAD Field Visit Report 3 points out, the process of setting up the VDCs is extremely
expensive and requires many hours of provincial and district staff. The project is well aware of
this problem and has taken steps to reduce this time period by eliminating the PRA and by
developing a much scaled-down version of data collection and needs assessment in setting up a
VDC.

The project is faced with a serious dilemma. There is much work to be done in existing VDCs to
make them more effective and sustainable. At the same time, government is preparing for the
1,600 Commune Councils to be set up in the year 2,000 and could request SEILA/CARERE to
become a part of that process. There is also much interest by the government in taking this
model to other provinces. Either program would require an enormous effort on the part of
SEILA/CARERE and would certainly mean that focus would move away from VDCs in their
five provinces to the new CCs in all of the provinces. Already, one SEILA/CARERE/LDF
province is testing out the Experimental Commune Program which eliminates the VDC

3
    “Cambodia, Field Visit report of February 1999 by ITAD”
UNCDF Cambodia Final Evaluation                                                           24


altogether and puts almost all focus on the CDCs. In addition, because there is less money
available to village projects during the second and third cycles, there is already a move by
SEILA/CARERE to focus more on the CDCs. This means that shoring up of the VDCs does not
seem a possibility at this time.

The question now is what will happen to the existing VDCs. If more attention is not paid to the
VDCs, they could easily become non-sustainable and simply fade away, particularly if donor and
government funds are not available for projects. This in the ET’s estimation would be
unfortunate, as VDCs with the LPP are giving Cambodia a first taste of a participatory local
government institution. On the other side is the potential for assisting the government to provide
an even more ambitious program in decentralized local government. It is quite a dilemma, and
before it can be answered, the government must make its desires known to SEILA/CARERE
regarding future expansion of the project into other provinces and its assistance in the Commune
Council process.


The ITAD report questions the sustainability and the replicability of the project citing the high
cost of organizing VDCs and CDCs, even though they agree that the intensity of the PRA
approach has been reduced and even the name has been dropped. The report says that “a
tendency to create complex, ambitious ‘international NGO’ procedures, systems, data-collection
and monitoring - - - which depend heavily on CARERE project staff and resources for
sustenance. The project staff is already addressing these issues by developing a shorter
orientation period, dropping the PRA, and using a much simpler data collection/needs-
assessment procedure. It has also moved away from the typical “community development”
approach and, in fact, is clearly stating that this is not a community development project.

The other more profound observation made in the report is that the project is not replicable in
other provinces. It states, “The evidence is that the outcome is a participatory area development
project rather than a replicable model for government and/or other donors.” Although the report
does mention some of the problems facing the government in Cambodia, e.g., The Cambodian
Government has to build its administrative capacity, including its ability to execute rural
development programmes, it appears to treat Cambodia in the same framework of other
developing countries that it is evaluating for the LDF, (Uganda, Malawi and Viet Nam). These
countries have little to compare with the devastation of infrastructure, human and institutional
resources and culture of Cambodia. Uganda has a strong president who is pushing perhaps the
most far-reaching decentralization process in the world. It has one of the finest universities and
some of the most educated civil servants in Africa. Malawi has had a stable government and
civil service for over thirty years. Viet Nam, although plagued with terrible war, never lost its
basic infrastructure, nor its human and institutional resources and culture. Therefore, if one takes
the position that Cambodia is somewhat different from other developing countries, then it may
be necessary, when attempting to implement meaningful developmental projects, to adapt some
distinct and unusual approaches to insure success. This is particularly true if the project deals
with local government decentralization, participation, capacity building and infrastructure
projects. If one does not accept this position, then there would be little argument with the report.
UNCDF Cambodia Final Evaluation                                                           25


It is argued that because so much is needed to bring Cambodia up to even minimal standards
found in other developing countries, particularly in human resources and institutional
development, that intensive capacity building programs have to be engaged. This requires the
staff intensive approach that the project has taken. When this training and capacity building has
been integrated into government staff and into the CDCs and VDCs, which is already appearing,
staff requirements will decrease and eventually will almost disappear.




       3.6.6. Results and Action- Oriented Management

Many of the positive results discussed above are greatly assisted by having specific projects to
complete with indicators, time-lines, budgets and clear means of monitoring. Staff must perform
within certain standards and in a time period. According to modern management theory, this has
a tendency to create a certain degree of challenge within staff, which in turn creates excitement,
and finally, good results are obtained. It is clear that people are focused on results and on the
action that brings those results. This is clearly different from that shown to the Evaluation
Team’s member during his training work in other provincial governments in 1996 - 07.


   3.7. Capacity Building

Indeed, from an institutional standpoint, the whole structure, as well as the process that has been
created from top to bottom, is impressive. It creates a structure and a process that facilitates
planning, communication (both vertically and horizontally), financial controls, transparency,
accountability, networking and a host of other excellent institutional activities.

One possible constraint is that there is no Ministry of Local Government at the national level to
oversee and coordinate the process of local government. While the Ministry of Interior is in
charge of all provincial offices, it must also involve other ministries such as Finance,
Agriculture, Education, Women’s Affairs and Rural Development in its local government policy-
making. As each ministry has its own interests to promote, differences can arise which will
delay progress on new policies and initiatives.


       3.7.1. Training

In one year alone, 56,245 people (34% women) were trained in 2,200 workshops in such things
as planning, problem solving, finance, water and sanitation, agriculture, management, gender,
governance, health, infrastructure design and implementation among others. Most of this
training was for specific job- related functions so that they could put the skills learned into
practice. Most important, however, was learning by doing and working with counterparts in
CARERE. The Evaluation Team talked with representatives of every level of government from
governors to facilitators and to people in the CDCs and VDCs. Everyone thought that the
UNCDF Cambodia Final Evaluation                                                           26


training had been extremely beneficial. At first, some of the trainings were too academic and not
properly related to the Cambodia situation. This apparently was corrected

Clearly, capacities have been built from top to bottom. Staff is using sophisticated planning
tools, implementing fairly complicated infrastructure projects and has learned an array of modern
institutional and management skills that are in many instances totally new. In fact, development
workers in Cambodia would argue that even the concepts are new. Such things as turning a
concept or an idea into a project are almost foreign notions to many Cambodians. Basic problem
solving as known in western countries is not a normal practice. There was evidence that this is a
common practice at least in the PRDC, EX-Com and Secretariat (in spite of the long conference
tables and seating arrangements).



   3.8. Lessons Learned

The project demonstrates the following:

1. The top management of CARERE had years of experience working with Cambodians in the
   field, knew their language, their culture, the horrors of their experience, and appreciated the
   enormous constraints to development that existed in the country at the beginning of
   CARERE 2.

2. Local government can be changed to be participatory and decentralized when there is “buy
   in” at the top of the provincial government. The present governors of both Battambang and
   Banteay Meanchey, where the project started, enthusiastically supported the mission of the
   project, as did other influential people in the provinces including two former governors.

3. A successful local government project in one province (in this case two provinces) can lead
   to replication in other provinces and eventually receive support by national government.

4. Although operating with the same mission and strategic objectives, there were different
   methods of fund allocation and project selection that allowed for considerable adaptation to
   local situations.

5. Intensive training at all levels of provincial government was followed by opportunities for
   staff to implement what they learned, while at the same time receiving back-up technical
   support. This process not only dramatically increased the capacities of staff but also changed
   their behaviors, attitudes and work habits and in the end, it greatly assisted in changing local
   government.

6. With a clearly defined “mission” by all the top players (What are “top players”? You had
   mentioned the problem of some people thinking they’re better than other people in
   Cambodia and here you’re talking about “top” players which implies that there are bottom or
   lower players) of the project, and through training and close supervision, the mission was
UNCDF Cambodia Final Evaluation                                                       27


    integrated among government staff. The project was therefore driven by a clear sense of
    shared goals and vision that was “bought” by the great majority of participants..

7. Beneficiaries who saw the possibility of immediate benefits from government realized from
   the beginning that the participation asked for was real and not another of the many empty
   promises from government.

8. Government was involved from the beginning and was the actual implementor of the project;
   therefore, the project had ownership throughout government.
UNCDF Cambodia Final Evaluation                                                           28




4.    THE LOCAL PLANNING PROCESS - FINDINGS

     4.1. Attainment of Objectives

The objective of the local planning process (LPP) as cited in the Project Document is to develop
a participatory, demand-driven rural infrastructure planning and implementation process to
support projects funded through the LDF that aim to generate greater economic development and
improve the quality of life for poor residents of rural areas.

As of August 1999, SEILA/CARERE/LDF has provided support to 1,144 locally elected Village
Development Committees (VDCs), 134 Commune Development Committees (CDCs), and five
Provincial Rural Development Committees (PRDCs). Each year as the CDCs gain experience
implementing the process, they demonstrate added maturity in needs assessment, planning, and
implementing projects, indicating that rural development structure is viable in Cambodia even
though the model is new to the context. The LPP makes an important contribution to
establishing two-way communication between administrative levels from a broad village base to
the province.

The LPP has also gained acceptance as a viable model for decentralized rural development with
possibilities for replication nationwide. Indications are increased collaboration and feedback
from NGOs working in the SEILA target communities, who initially had reservations about the
model, and the decision by donor agencies to support the LDF.



     4.2. Planning Process and Cycle

        4.2.1. Planning Process

As shown in Project Document section 2.5, the LPP begins the six step process at the local level
where villagers collect socio-economic data, identify needs, prepare village maps and develop
village future action plans. A simplified exercise is conducted after the first year to update the
plan. Changes being introduced to the process are discussed below.

These plans are taken to the commune level where the CDC prepares 3-year Commune
Development Plan and a list of projects they wish to implement the following year. Departments
provide communes with sectoral policy guidelines to inform their planning. These are introduced
to the CDCs during the preparations for the District Integration Workshop. For example, the
Ministry of Health does not provide health centers for communes with less than 10,000 residents
and communes need to keep this in mind when formulating their plans.

At District Integration Workshops, communes present their plans and the departments provide
feedback on what resources they have to offer. It is at this time that priorities are discussed,
resources are allocated by sector and the communes extract commitments from the departments.
All departments receive copies of the commune plans before the workshop so they come
UNCDF Cambodia Final Evaluation                                                                29


prepared to respond. They also receive basic data for their sector from the commune database.
NGOs, IOs and other agencies are to attend the workshops and present their activities.

       The Evaluation Team had the opportunity to observe a number of District Integration
       Workshops and was impressed by the lively discussions that took place. There was a
       noticeable difference for districts with more experience: districts in Banteay Meanchey,
       for example, have cut the workshops down to one day and eliminated long speeches,
       getting straight down to the nuts and bolts. Eighty people spent most of the day sitting on
       the floor in sectoral working groups discussing -- even arguing -- the merits projects,
       excessive expectations, resolution to agreements from the previous year that had not
       been honored, and ways to combine resources to get projects done. The atmosphere
       was buzzing with energy and in the midst of it all, the District Chief and Department
       Directors were beaming. They were listening to the communes, and they had something
       to offer them. Certainly not the average government meeting…


After the workshops, the CDCs are then responsible for allocating their own resources, taking
into account the letters of intent signed between CDCs and Departments or other organizations at
the District Integration Workshops. At that time they receive an indicative planning figure (IPF)
from the PRDC to use as an estimated budget. The purpose of this is to allow communes to
practice allocating resources within a fixed budget, rather than writing out a simple wish list.

CDCs then formulate project proposals with assistance of the provincial technical support staff
(TSS) who the communes to assess project feasibility and estimate costs. The formulated
projects are forwarded to the PRDC Secretariat for review and submitted to the PRDC Ex-Com
for approval.

Asia Institute of Technology (AIT) assists with a figure describing the process on the following
page with figure 1
UNCDF Cambodia Final Evaluation                                                               31


process. VDCs are effectively guessing, admittedly strategically, about the projects most likely
to be approved. As a result, they may be proposing projects that are not their top priorities.

       Using participatory feasibility studies might have solved this very problem in Doung Treng
       village of Preah netr Preah district. Villagers there keep insisting on a large and
       expensive bridge, but have had their project rejected each year. Although have saved
       money for a local contribution 2 years in a row, each time they have used the money for
       something else. Now the village is trying for the same bridge in year 3 but with no
       savings for their local contribution. In discussion with villagers and the CDC Chief, they
       did not want to consider other options for smaller bridges that might have been more
       affordable and still solved their transport problems. A participatory study may have
       helped them to see how other options were also feasible.

The Provincial Department of Planning then finalizes a Provincial Development Plan that
includes sector plans, as was an investment plan for the province. These are forwarded to the
national level where they can inform national planning.

One of the lesser-known aspects of the program is the provincial CDC/VDC congress. The first
was held in Battambang in December 1998. A second was held in Banteay Meanchey in March
1999. In Battambang 560 village and commune representatives attended, while in Banteay
Meanchey 1,200 attended. The purpose of the congresses was to bring participants together for
two days to inform them of the LPP process, share experiences, highlight the outputs and
stimulate networking between VDCs/CDCs. In addition to representing an astounding logistical
feat, the congresses most certainly represent a first for Cambodia, and their potential for other
dissemination and communication activities in the light of the upcoming commune elections
merits further attention.

       4.2.2. Planning Cycle

The development of village and commune plans takes place in the rainy months between June
and August, so that completed development plans can be presented at the district integration
workshops in September. The allocation and formulation of projects takes place at the commune
level in October and November, so that the projects can be reviewed and approved at the
provincial level by the end of December.

As rural life in Cambodia is inextricably linked to the coming and going of the rains, the timing
of the planning and implementation cycle has proved to be important. In order for projects to be
implemented during the dry season, from January to April-May, it is essential for project
approval to be completed by the end of the year.


       4.2.3. Changes to the LPP

Over the past three years, the LPP has been closely monitored and several evaluation workshops
and reports have been made. As a result, the process has been simplified and the number of steps
involved reduced to 21. While at first glance this sounds like a lot, all the steps are needed to
ensure a comprehensive system of feedback, consensus building, monitoring and approval.
UNCDF Cambodia Final Evaluation                                                            32


Content and precise number of steps differ between provinces according to how activities
deriving from the agreed upon LPP standard are organized and implemented.

Most of the changes in the LPP have been related to streamlining the process while developing it
as a model that can be replicated. The most significant ones relate to the bidding process, the
simplification of the data collection system and added focus on building the capacity of the CDC
members.

The most important change in the LPP structure has been the introduction of the bidding process.
This process brings in contractors as a third party to program implementation, with the CDC as
the ‘owner’ of the project, and the Ex-Com/TSS monitoring technical quality and authorizing the
disbursement of funds upon completion of each phase of the project. This tri-partite structure can
be viewed as practicing the separation of powers at the grassroots level and promoting good
governance.

Another change has been the simplification of PRA tools, which proved to be too complex as
introduced initially. These have been simplified into what are now referred to as data collection
tools. This change has reduced time and effort demands placed on villagers to complete tasks not
directly related to decision-making that can tax the interest of the villagers. As mentioned above
in the section on participation, however, placing additional focus on activities that do encourage
greater participation in decision-making by villagers has yet to be done.

Finally, since the beginning of the project, added emphasis has been placed on developing the
capacity of the CDC members, encouraging them to do more of the work themselves (such as
formulation of project proposals) and facilitating communication with the commune via radios.
The result of these efforts was evident in the noticeably more advanced communication skills of
members of CDCs in their third and fourth cycles compared to those of more recently elected
CDCs.


   4.3. Resource Allocation

       4.3.1. Selection of Communes

The first SEILA target areas were carried over from the initial phase of CARERE. When the
pilot LDF program began in 1996, three districts in Battambang and two in Banteay Meanchey
were still under Khmer Rouge control, in addition to portions of other districts. Due to security
concerns, which lasted as late as December 1998, the initial selection of communes was done
using the criteria of security and accessibility. Thus in Battambang and Banteay Meanchey the
first communes to access the LDF were usually located closer to the provincial centers or in
districts most removed from the fighting.

As security and accessibility improved, the criteria for selection expanded to include populations
with a high prevalence of returnees or internally displaced people, contiguity with existing target
communes, likely effectiveness of investments, visibility of demonstration, the existence of other
IO and NGO programs in a district, and cooperation of local authorities.
UNCDF Cambodia Final Evaluation                                                           33




Up until 1999, there were no sources of disaggregated data available, so commune selection
could not be based on poverty indicators. Now that the Commune Database is available,
comparisons between villages and communes are possible. Selection of new target communes is
now based on combinations of poverty and population data, with some variation of methodology
between provinces as discussed below.

With the reintegration of the last Khmer Rouge strongholds into the government administrative
structure, between May 1996 and December 1998, the development community implemented an
organized response to provide emergency assistance to the former Khmer Rouge districts, known
today as the reconciliation districts. SEILA began implementing the LPP soon afterwards,
beginning with two communes in each reconciliation district.

       4.3.2. Allocation of Indicative Planning Figures to Communes

The indicative planning figure (IPF) is a budget allocated to each commune from the provincial
level. Communes then allocate their IPF resources to specific village projects through the local
planning process, based on the merit of proposals received. There are no requirements to
distribute or concentrate the resources to a large or small number of villages.

LDF policy as per the original project document was to allocate $50,000 to each of the
participating communes over the three year cycle, regardless of the number of villages, their
population, or need. The general principle was to disburse $25,000 during the first cycle,
$15,000 during the second cycle and $10,000 during the third cycle. There were no specific
criteria used to adjust these amounts.

From this basis, allocation methods have evolved over time and varied between provinces, using
various combinations of poverty indicators and population. In the second year of the LDF, a
reassessment of allocation methods took place, based on internal reviews and external
monitoring. It was agreed that a flat rate per commune was not appropriate and that the IPF
should be adjusted in some way to better reflect "need". Initially, only the number of villages per
commune was used. Poverty indicators were added later.

Banteay Meanchey was the first province to use objective poverty criteria as well as indications
of support from outside organizations. Battambang has followed this method beginning in 1999.
The formulae differ between provinces but in general use a weighted average of the poverty
ranking, derived from the Commune Database, and the population. There is also some discretion
available to PRDCs in the allocation process.

As a result of this evolution, the actual amounts allocated to communes have varied
considerably, in some cases double or half the average amount over a three-year funding cycle.
By the end of the third year of the LPP cycle, communes should have received about $50,000.
The actual amounts allocated for these communes range as high as $96,000 in Battambang for a
commune with 18 villages.
UNCDF Cambodia Final Evaluation                                                                      34


The indicators that are used for IPF allocation from the provincial level to communes need
improvement if they are to be used for targeting. To indicate poverty, Battambang Province uses
a combination of children not attending school; the amount of poor housing; and the proportions
of invalids, returnees, internally displaced persons (IDPs), and female-headed households.
Banteay Meanchey uses almost the same set of indicators, but replaces IDPs and returnees with
landless households. There are several problems with this set of indicators.

First, some of the indicators are poor proxies for poverty. The percentage of children attending
school is also strongly affected by the proximity of schools and by transportation problems. The
most widely accepted poverty indicator, the number of female-headed households, is also not a
good indicator, as there is evidence that these households are in fact better off than are male-
headed households. 4 Furthermore, an analysis of the variables contained in the Commune
Database shows that the proportion of female-headed households has no relationship to the
proportion of children in school, vehicles per family, or household equipment. This may be due
to these households being smaller through a truncation of fertility, and it may reflect the role
women have as traders.

Second, the variables in any set of poverty indicators - if they are accurate – are colinear by
definition and therefore do not all need to be included. Any single indicator of consumption, for
example, is as good as any other since they will all reflect relative household income and
therefore the underlying incidence of poverty. Vehicles, housing condition, and televisions are
probably all equally good indicators of consumption.

Third, the indicators that reflect personal status during the political upheaval of recent years may
not remain accurate indicators of poverty. Returnees and IDPs would surely have been
disadvantaged in the early stages of their resettlement, but that disadvantage probably weakens
over time. Many returnees have been able to leverage their knowledge of English to gain
employment for example. If the point of these indicators is to say something about access to
farmland, and by implication farming income, it would be better to use the variables related
directly to land.

A correlation analysis of the poverty indicators currently in use is reported in Annex VII. The
results show that the only moderately strong relationships are those between housing quality and
the numbers of motorbikes and cars per family. Other commonly used indicators have no
relationship to landlessness, refugee status, or school attendance. These results suggest that
consumption-related items are probably the most consistent indicators of relative wealth. 5

There is widespread agreement that simple criteria work best and are most likely to be applied
consistently. The capacity of many of the staff involved in the LPP to apply quantitative
formulae is limited. Since there are consumption related items in the Commune Database, they
should be used in lieu of more complicated schemes. Using a poverty criterion alone, however,
would imply that the same villages should always receive the largest allocations, and therefore it
makes sense to include population so that allocations are weighted fairly for larger places.

4
  Ministry of Planning. Cambodia Human Development Report 1998. Phnom Penh, 1999
5
  We assume that the "Wealth" field in the Commune Database, currently empty for all places, would be derived
from existing variables and does not represent additional survey information.
UNCDF Cambodia Final Evaluation                                                             35




       4.3.3. Allocation to Projects

The original intent of the project allocation process was to use a multi-criteria framework,
developed by UNCDF, AIT, and CARERE that included criteria of project cost, number of
beneficiaries, priority of problems or needs, cost effectiveness, time frame, local contribution,
and sustainability.6 This framework has proved overly ambitious in the Cambodian local
context. Interviews and monitoring reports (both internal and external) have revealed much
confusion at the provincial and commune level as to the process used for allocation. 7 CDCs have
tended to use only a few criteria in practice, usually the ones they think they understand.

Monitoring reports and LPP review reports attempted to resolve the difficulties encountered in
applying the ranking and screening criteria. However, CDCs continue to have problems. The
1999 Independent Evaluation of the local planning process observes that some CDCs continue to
rely on the facilitators for the allocation of the LDF and recommends that there be only two or
three simple criteria given to the CDC members to guide them in their decision making. The
latest review of the LPP has noted that the Commune Database is now available, and the
consultant responsible for Reconciliation programs has recommended simplifying the range of
indicators. The Evaluation Team acknowledges that CARERE management is aware of the
problem and that it is on the agenda for the 1999 Battambang LPP Review Workshop.

Allocation to projects should therefore be simplified. The number of beneficiaries and some
prioritization of need by the CDC may suffice. An example can be taken from the Vietnam LDF
project. At the equivalent stage of the LPP, villagers rank their perception of the need of every
other village besides their own to determine the neediest village(s).

However, continued experience with the allocation process has increased participants’
confidence in the process, lending support to the general impression of capacity building among
participants at all levels of the LPP. Despite confusion, the CDCs are believed to allocate
resources to the highest-priority projects that are proposed by VDCs. The fact that IPF
allocations are reduced after the first year has forced CDCs to make tougher decisions, and led to
selection of fewer projects, more often at commune rather than village level. Most importantly,
CDC and VDC members interviewed during the mission stated that they believed a rational
quantitative method was useful as it enabled them to be impartial and "fair."

Commune Development Committees are not yet effective in the prioritization of project
proposals, consolidation of village proposals into larger commune projects, or identifying wider
needs.8 Between villages, the IPF has typically been allocated on an equal opportunity basis,
according to “one village, one project”. In Battambang, at least 96% of IPF funds were allocated
to village level projects in 1997 and 1998. In Banteay Meanchey all funds were allocated to

6
  Pongquan , S. “Monitoring and Evaluation Mission on Commune Development Committee IPF Allocation
Process (17-28 March 1998)”, Cambodia LDF Report 24, Bangkok: Asian Institute of Technology, April 1998
7
  Biddulph 1999a, Biddulph 1999b, Pongquan 1998, M&E and LCB Staff 1998
8
  M&E and LCB Staff, “Study of the CDC IPF Allocation in Battambang and Banteay Meanchey 1998”, CARERE
report LPP/-/CAR/666
UNCDF Cambodia Final Evaluation                                                               36


village level projects over the same period. However, several CDCs visited by the Evaluation
Team, that were in their third year of the LDF program, said that they could see the benefits of
combining smaller projects into larger ones, and were beginning to discuss ways to do so.

However, these figures do not tell the whole story. In some cases, larger projects at commune
level have been split up and administered as village projects. The reasons for splitting up
projects include the administration fee, collection of local contributions, and labor organization.
Therefore, the proportion of commune-level projects cannot be accurately detected in the LDF
database, but is certainly higher than zero to four per cent.


       There have been some notable exceptions to the trend of "one village, one project". In
       Phnom Sampou Commune in Battambang, a canal that was identified as a top priority for
       one village would also benefit 4 other villages. The 5 VDCs agreed to combine their
       resources and put the project forward as a CDC project. In another case, a 9.3-km road
       in the commune was built under the SEILA Program for sectoral support using ILO
       methods. It is now maintained by people who live along the route, even in the long empty
       stretches between villages. VDCs arrange maintenance using unpaid workdays. This
       area is a prime orange-growing region and people value the road for market access. So
       they have set up checkpoints that allow them to collect tolls from trucks using the road,
       and to close the road during the rainy season when needed. Materials for repairs are
       bought using tolls, and the provincial authorities have helped by soliciting contributions
       from the big trucking companies that are active in the area.




   4.4. Participation

       4.4.1. General Participation

The goal of the local planning process is to strengthen the ties between state structures and civil
society by promoting broad-based participation in the development process to access the LDF.
This is a daunting challenge given Cambodia’s recent history, the high poverty rate and planning
process has achieved active participation at the grassroots level by encouraging villagers to:

       •   Participate in village meetings
       •   Select the VDC members
       •   Collect and analyze socio-economic data
       •   Identify needs and priorities in order to prepare a Village Future Action Plan
       •   Contribute to the construction of infrastructure projects in the form of labor, supplies
           and cash
       •   Maintain the projects

By the end of 1998, 5,705 villagers were participating in the VDCs in the five provinces.
However, the 1999 external evaluation of the LPP (Biddulph, August 1999), which concentrated
on participation indicates that outside the data collection/village mapping exercises and
community contributions, villagers have little involvement in the LPP. VDC meetings are held
only when there is a project involved. (there may be a better word than “involved.” They are
UNCDF Cambodia Final Evaluation                                                             37


often not well attended. In addition, the meetings usually require a government facilitator to
conduct them. After a period of time, some members lose interest, leaving only two or three
members actively involved.

The evaluation report defines participation as the influence that villagers have over development
whether in the form of direct participation in decision-making or of representation through
elected representative who make decisions and work on behalf of their constituency.
The report finds that almost no villagers are aware of the commune development fund or of the
commune committee and rightly points out that access to information is the key to initiating
accountability. If villagers are not informed about the LDF, there is no hope of their being able to
influence the process. If they are informed about the project, traditional methods of
communication and of pressuring local authorities have a chance of being effective.

Two important recommendations have been made to address this issue. The first it to improve
the effectiveness of the Village Information Strategy to raise awareness about the public’s rights
and responsibilities (and those of their village representatives). The second is to use village
meetings to promote the engagement of the community in dialogue relevant to the LPP decision-
making process, specifically after the CDCs meet to decide on projects but before the project
proposals are formulated. It is at that time that villagers can have a real say in who should
manage a project and where a project, such as a well, should be located.

       4.4.2. Role of Poor

The goal of the LPP and the LDF is to generate greater economic development and improve the
quality of life for poor residents of rural areas. The main area in which the poorer sections of the
population participate in the LPP is during the village meetings at the start of the cycle when
priority needs are identified.

An in-depth assessment of the participation of the poor in the VDCs and CDCs was not possible
in the scope of this evaluation. However, it is unlikely that they are represented to any great
extent, first because of their own time constraints as they seek to earn a living and second
because the impression that committee members should be educated is a barrier to participation.
A number of characteristics limit the opportunities for targeting participation of the poor over
any other group in the LPP. As a mechanism for decentralized planning and financing, the focus
is on establishing a system of interaction between government and civil society. In order for
vertical integration to occur, points of communication have to be established, and this necessarily
limits broad-based participation. Second, as most of the projects are infrastructure projects such
as road, the poor are most certainly benefiting from them. Finally, as the general level of poverty
in Cambodia is so high, targeting the poorest sectors at the village level is not necessarily
meaningful. The 1999 SIDA Advisory Team report confirms that while the bulk of LDF
beneficiaries qualify as poor, “CARERE/SEILA does not reach the poorest strata and is not
designed to do so (p. 8).”

While effective targeting of the poor at the village level is limited, the opportunities are much
greater at the provincial level. It is during the selection of communes and setting of commune
allocations that the SEILA Program can have the greatest impact on the poor. With the
UNCDF Cambodia Final Evaluation                                                           38


establishment of the Commune Database in early 1999 and the review and improvement of
poverty indicators, SEILA management will want to be tracking PDRC’s commitment and
ability to target poorer communes.

          4.4.3. Role of Women

Observers at a village meeting will note that the majority of participants are women. This is true
of most village meetings, as the men are usually busy working in the fields. However, when it
comes to representation and decision-making in Cambodia, the women quickly become
sidelined. The requirement that 40% of VDC members be women has been an important step to
incorporate representation of a majority of the population that traditionally stands outside the
village and commune decision-making.

The SEILA Program supports the Ministry of Women’s and Veterans affairs at national,
provincial and district levels. Between 1996 and 1998, more than 130 ministry staff attended
training on gender awareness. In turn the Provincial Department of Women’s Affairs staff
conduct gender workshops for the VDCs and CDCs and assigns a Gender Focal Point who
receives additional training.

Gender Focal Points have also been assigned in the departments, although their participation in
meetings with the Department of Women’s Affairs is low. A dilemma exists as to whether the
department gender focal point should be high ranking or low ranking as high ranking officials
are not liable to go to the meetings, and low ranking officials are not able to influence policy
within their departments.

Other policies, such as the requirement that a gender assessment be completed for every project,
are helping to bring women’s issues into the limelight. At the District Integration Workshops,
discussions on women’s health and domestic violence are held for all to listen and participate in.

Despite all these efforts, program mangers, staff, and gender focal points agree that the
participation of women in the LPP is still low. Specifically, they are not active in decision-
making with regards to LDF allocation and their presence at the commune level remains limited.
At the end of 1998, only 21% of CDC members were women, below the 30% requirement. At
the district level, the figure drops to 8%. Among government staff at the provincial level,
women continue to represent a minority, only 19%.

One of the main constraints to grassroots women’s participation is lack of confidence. The 1998
LPP external evaluation indicates that women feel shy about participating and believe that
members of the VDC should be well educated. However, participation increases as the
proportion of women within a group increases, highlighting the importance of efforts to increase
the proportion of women represented at the commune level.

Another important barrier to participation is lack of time: as mothers and often heads of
households (25.3% nationwide),9 it is difficult for women to find the time to participate actively
on a committee. Being on a CDC requires more time and the woman may have to walk a few
9
    1998 Cambodia Human Development Report, Ministry of Planning
UNCDF Cambodia Final Evaluation                                                             39


kilometers to attend a meeting. It is not unusual for women members to become inactive or drop
out of the committee when domestic demands on their time increase, according to the season.
This constraint cannot be lifted, however, and the program can only continue to actively
encourage participation.

However, exceptions do exist and with ongoing efforts, participation of women should continue
to increase. In Banteay Meanchey, for example, at least a dozen VDCs are made up of a majority
of women, 4 have four women members, and one is solely made up of women. Also in Banteay
Meanchey, the Director of the Department of Women’s Affairs was a Commune Chief for four
years and a District Chief for four years. Although far more needs to be done, the project has
made a major step in bringing the role of women into the local government/decentralization
process


   4.5. Attitudes of Staff, Officials, Community

The evaluation team was impressed by the attitudes of the CARERE and SEILA staff, including
PRDC officials. The energy level in the program is high, and people are enthusiastic despite the
large workload. The three PSO Managers interviewed are generally pleased with the level of
support and open-mindedness of government officials. It is clear that the success of the program
depends to a great extend on the provincial government leadership.

Encouraged by CARERE, government officials have taken important steps towards improving
performance. For decades the Cambodian civil service has been a stagnant bureaucracy
reinforced by a traditional culture of tolerance that does not tend towards measuring performance
and removing staff when they are found incompetent. The rural structure has been further
ossified by the long tenure of district and commune chiefs who were for the most part appointed
in the early 1980s.

As the program has gathered momentum and begun to show the fruits of all the efforts that have
been put, in the PRDC leadership in the three provinces visited has shown itself willing to break
away from the traditions that have burdened the government bureaucracy. A rigorous
performance evaluation of DFTs, PFTs, and TSS was conducted in 1998 and a number of staff
were returned to their line departments. The PRDC leadership has been willing to recommend to
the Ministry of Interior that Commune Chiefs be replaced when their management styles was
perceived to be blocking the LPP and participation in the development process.

At the district and commune level, officials have received the SEILA Program with open-
mindedness. District officials are pleased with the district integration workshops attended by
district officials, provincial and district department staff, and commune representatives. Villagers
indicated that over time commune officials participating in the LPP are more accessible to
villagers and more inclined to go out into the communities and listen to what villagers have to
say.

The success of the SEILA Program in effecting changes in attitudes, motivation and capacity of
government staff highlight the need and importance of civil service and administrative reform.
UNCDF Cambodia Final Evaluation                                                           40


Sustainable change, within the SEILA Program and in general, cannot take place within the
government context until technical and financial resources are made available to government
staff. The SEILA Program makes it clear that when they are, much can be accomplished.

   4.6. Collaboration

Collaboration with other agencies in the LPP takes place at the central level in the form of
Memoranda of Understanding with agencies such as the World Food Program. Negotiations are
currently underway with the Social Fund, a World Bank project that also supports rural
infrastructure projects.

NGOs are a natural complement to the LPP process as they play a role providing skills and
resources to communities and they have more experience implementing non-infrastructure
projects. NGOs are included in the LPP during the District Integration Workshops, where they
too present their plans for target areas and interact with the CDCs to discuss possible areas for
collaboration.

NGOs interviewed by the EvaluationTeam that have close collaboration with the SEILA
Program were very positive about the program. They cited added facility in organizing sub-
committees, such as credit groups, as one of the benefits of working with the VDCs.

As NGOs in general initially viewed the program with skepticism, this is a testimony to the
success of the program. However, it is the missed opportunities for collaboration that are
noticeable in the number of NGOs that are not familiar with the program (usually the ones who
are skeptical). While the Evaluation Team recognizes that SEILA has already made efforts in
this direction, more efforts are needed to reach out to the NGOs and help them understand how
SEILA and NGO resources can be pooled together, notably in strengthening the skills of the
VDCs and CDCs.

Within the program, SEILA has identified collaboration with monks and traditional leaders as
opportunities for strengthening the LPP and is already working towards increasing relationships
with these groups.


   4.7. Assessment

       4.7.1. Implications for good governance

The success of the LPP model has direct implications for good governance. Participants at the
grassroots level learn to articulate their needs, allocate resources and practice accountability,
important prerequisites for good governance. Government officials also learn facilitation and
planning skills, and practice listening to and negotiating with the CDCs.

The interaction between government and civil society, in a sense the heart of the LPP, occurs at
the annual District Integration Workshops. The workshops allow commune representatives to
UNCDF Cambodia Final Evaluation                                                                   41


interact with provincial department officials, learn about department policies, and negotiate the
use of resources allocated to the departments.

The SEILA Investment Program (SIP), which makes resources available to the government
departments, plays an important role in the success of the LPP even though it falls outside the
LPP itself. Without these resources, government officials would be empty-handed and there
would be no reason for negotiation.


       4.7.2. Capacity of VDCs and CDCs

Once the new VDCs and CDCs have been selected in a commune, they are brought together to
the commune center for three days of training for an overview of the LPP and basic concepts of
community development. The also attend various training workshops organized for them by the
SEILA Program, including human rights education and gender awareness,

Starting capacity of the VDCs and CDCs is usually very low, as they have little education and no
previous experience using data collection tools and/or developing plans. As a result, facilitators
sometimes take a dominant role in the local planning process.

The capacity of VDCs and CDCs increases significantly as they go through more cycles.
Notable changes include improved communication and analytic skills. From one year to another,
changes in commune chief presentation skills at the district integration workshops are marked as
they move away from simply reading off a flipchart to facing the audience and providing
explanations. As CDCs mature, they are able to provide orientation to new CDCs in a commune-
to-commune approach being tried in Siem Reap.

CDCs also gain more confidence, daring to move away from allocating one project to each
village to looking at overall commune needs. Confusion over the allocation process continues to
be cited as a problem area. This could be reduced by the provision of simpler guidelines and
limited criteria.

       Improvements in the capacity of CDCs and VDCs should increase their legitimacy. This
       is important for the committees to have the support of villagers for development planning.
       In Doung Treng village of Preah netr Preah, villagers and the CDC Chief stated that if
       they did not get the bridge they wanted through the LDF, they would form a “special
       committee” with better links to relatives overseas and in the cities, to build the bridge from
       private donations. They said the VDC could not do this effectively. Clearly the VDC in
       this village is only a means to LDF funds, not a generic development committee.

Planning tools used in the LPP include village maps drawn on the ground, commune maps drawn
on flipcharts, and tables used to prioritize needs and projects. These tools are simple, but they are
appropriate to the context and represent large strides in conceptualization and analysis for the
participants. New committees have more difficulty with the concepts and tools and tend to rely
on the facilitators for assistance. CDCs that have been through two and three cycles are much
more comfortable with the tools and they are generally considered able to do adequate planning
and design projects that meet the priority needs of the population.
UNCDF Cambodia Final Evaluation                                                            42




       4.7.3. Sustainability of the LPP Structure

The LPP has potential for institutional sustainability. The government staff manages the
organizing of the VDCs and CDCs elections, and subsequent training and planning activities.
District representatives are enthusiastic about the integration workshops and play an active part
in them.

Program staff continuously seeks new ways to enhance learning. For example, in Siem Reap, the
districts include non-target communes in the District Integration Workshops in order to give
them some exposure to the process. Non-target area commune chiefs interviewed at two district
integration workshops were enthusiastic about learning about the process, although naturally
desirous of having access to resources too.

Concerns about the sustainability of the LPP structures center on:

•   In-house capacity of the facilitation team to upgrade their facilitation skills and train new
    facilitators (see capacity-building of VDCs above);
•   Current insufficient staff to cover an increasing workload and attain full provincial coverage;
•   Potential institutional memory loss after a new round of VDC elections.


       4.7.4. Focus on Commune Level

SEILA management is shifting the focus of the program to the commune level. This makes sense
in light of limited resources and the expectation of an upcoming Commune Administration Law,
but has important implications for participation. As it is now, villagers have little knowledge of
commune development committees and the allocation process. This situation will be reinforced
in the model currently being tested in Siem Reap that replaces the VDC with two village
representatives, narrowing the lines of communication within the village.

In fact, not all "village" projects are isolated efforts. There are cases of larger project that might
appropriately be called "commune" projects being split for funding purposes into village-level
projects. This is because VDCs receive a contract administration fee and are responsible for
collecting the local contribution and for organizing labor.

In preparation for the next phase of SEILA, the team recommends a review of the long-term
objectives of the LPP with regards to village participation and a clarification in policy to avoid
unreasonable expectations with regards to participation.
UNCDF Cambodia Final Evaluation                                                                  43


       A limited number of villages have maintained a practice initiated under the previous
       socialist regime of community rice fields that are distributed each year to villagers through
       a lottery system. When the harvest is reaped, the farmer pays a fixed amount to the
       village. In Ankeal Boh Village, Koh Pongsat Commune in Banteay Meanchey, for
       example, the 10 hectares of community rice fields provide an income of 20,000 Baht
       (US$526) per year to the village. This money was used for the community contribution to
       the LDF project. This model could have important implications for future community
       activities with the enactment of the Commune Administration Law, which is expected to
       give communes the right to raise their own resources.
UNCDF Cambodia Final Evaluation                                                                   44




5.     Findings: project design, implementation, and outputs


The institutions involved in the SEILA Rural Development Structure (Village Development
Committees, Commune Development Committees, and Provincial Rural Development
Committees) have overseen about 1,300 local projects valued at about US$3.8 million. Local
support in the form of cash contributions has totaled US$500,000, or 16% of total project costs.
Projects completed under the LDF represent approximately 8% of total CARERE activities.


     5.1. Institutional support and incentives

The Royal Government has very little ability to finance development activities and is primarily
dependent on donor funding to implement it development plans. The extent to which the LDF,
in the context of the SEILA program, can serve as a model for a universal decentralization and
planning policy relevant to both donors and all levels of government is very important.10 There
is evidence that this is indeed happening.

In 1999, the Royal Government made a significant contribution to the LDF of US$215,000. The
Government has now announced it will double this amount, to US$500,000, for the year 2000.
Other official support has recently been announced by Ratanakiri province, which has decided
that line departments contribute 10% toward LDF projects, and Battambang province, which has
committed to contribute 20% toward a road maintenance fund.

Other donors too have contributed to the LDF. The World Bank has added US$500,000 million
to the LDF for reconciliation programs (and another US$1.1 million to other programs through
the SEILA financial system). AusAID is also contributing to reconciliation programs through the
LDF, with A$500,000 going to Pursat and Battambang provinces through the SEILA financial
system. IFAD is using the LDF to channel US$8.6 million over five years into existing SEILA
provinces. These sources of support will form the basis for the proposed Decentralized
Development Fund (DDF) that will eventually be used to channel funds from other sources.


     5.2. Delivery of rural infrastructure


        5.2.1. Sectoral allocation

The sectoral allocation of LDF funds has overwhelmingly favored transportation works, which
make up more than 40 per cent of the total. Irrigation, water/sanitation, and building (mainly
schools) projects make up the balance. Most of the non-infrastructure projects (mainly rice
banks and buffalo banks) are in Pursat and Ratanakiri, where they account for 43 and 74 per cent

10
 Charny, J. “Issues for Decentralized Planning and Financing of Rural Development in Cambodia”,
UNDP/CARERE, ca. 1998
UNCDF Cambodia Final Evaluation                                                                       45


of the totals respectively. The table is derived from provincial LDF databases that were updated
in September 1999. Detailed breakdowns of provincial allocations by year are shown in the
Annex VII.
                                                     Table 2

                          Sectoral Allocation of Projects, all provinces, 1996-1999

                 Type             Number        % of Total     Total Cost % of Total   % Local
                                                 Number          US $       Cost     Contribution
         Buildings                   162            9             582,511    16          15
         Irrigation                  161            9             305,650     8          18
         Roads / bridges             695           40           1,673,308    45          17
         Water / sanitation          459           26             640,332    17          17
         Non-infrastructure          279           16             522,335    14          13
         Total                      1,756                      $3,724,135

One possible explanation for the large proportion of road projects is the fact that villagers were
able to implement labor-based road construction themselves, in effect generating local
employment. Now that VDCs can no longer pay local labor in cash, the proportion of these
types of projects has fallen.

The sectoral breakdown of projects has been questioned in some provinces. Non-infrastructure
projects such as rice banks, agricultural machinery, and schemes to provide draft animals are
potentially the most controversial with respect to their nature as either private or public goods.
There is always the possibility that these projects will become private property, or at the least are
monopolized by a few people. There is in fact some evidence that this has begun to happen in
Cambodia.11

The Evaluation Team was not able to observe any non-infrastructure projects during the mission.
However, based on evaluation reports and experiences in other countries, these do not seem the
best targets for LDF funding. Small-scale economic development and micro-credit schemes
such as these require an entirely different planning process and project management structure
than are present in the Cambodian LDF. Rural infrastructure is not directly productive by
definition, and is well suited to an annual representative planning process. The kinds of non-
infrastructure projects that have been pursued in Pursat and Ratanakiri need continuous
management, guarantee schemes, and participation incentives. Without them, such projects are
likely to fail (in the case of credit) or become private property (in the case of machinery and
animals). If this happens, the long-term success of a participatory decentralized planning system
will be jeopardized.

The district integration workshops are working to improve the sectoral allocation of all sources
of funding, by making line departments consider work plans and sources of support for a
commune at one time. Sectoral targets for types of infrastructure exist in all the line
departments. For example, there are official standards for the number of classrooms per capita,
or the number of wells per household. The PRDC have a role in integration at the provincial
level through their oversight responsibilities, and this role should be exercised as early in the

11
     Biddulph, R. Independent Monitoring and Evaluation of the SEILA Local Planning Process: Final Report , 1999c
UNCDF Cambodia Final Evaluation                                                                46


planning process as possible. The annual production and review of Commune Development
Plans would be a good time to reinforce project approval criteria and guide CDCs and VDCs in
project selection. Since these responsibilities will increase with the establishment of a
decentralized development fund, such activities under the LDF will provide a good basis for
proactive sectoral planning.


        5.2.2. Cost Effectiveness

The introduction of bidding procedures has led to significant cost savings on village level
projects.12 As of mid-1999, nine of eleven communes in Battambang Province had carried out a
bidding meeting. All other provinces are implementing bidding this year as well. Cost savings
after bids average from zero to about 6 per cent of the estimate in Battambang depending on the
category of project, as high as 10 per cent in Banteay Meanchey, and up to 21 per cent in Siem
Reap. Although this comparison is only to the engineer’s estimate, it is the best indicator
available of cost effectiveness. A summary of cost savings by project type is shown in the table
below.

                                                 Table 3

                     Project Cost Savings Resulting from Competitive Bidding

                                  Type              Number      Total Cost   Savings %
                   Battambang
                                  Buildings                 4      16,085           -3
                                  Irrigation               10      14,944           -2
                                  Roads                    74     168,402           -4
                                  Water                     3       4,029           -9
                                  Non-infra                 5      46,865            0

                   Siem Reap
                                  Buildings                 6       9,500          -21
                                  Irrigation               12      12,151          -13
                                  Roads                    40      84,548          -11
                                  Water                    36      31,479           -9



Comparable data on construction costs from other providers were not available during the
evaluation mission. Provincial departments of public works were unable to provide cost data to
the Evaluation Team, and the private sector contractors that were contacted did not have enough
experience to describe “typical” unit costs. Unlike Vietnam, standard rates for public-sector
construction do not exist in Cambodia, so there is no published basis for comparison.

The Social Fund of Cambodia provides the closest comparison in terms of rural infrastructure.
However, Social Fund data are not given in unit cost terms (e.g., dollars per kilometer of road)

12
  Biddulph, R. “Battambang Province Feedback Report”, Independent Monitoring and Evaluation of the SEILA
Local Planning Process, 1999a; Biddulph, R. “Banteay Meanchey Province Feedback Report”, Independent
Monitoring and Evaluation of the SEILA Local Planning Process, 1999b
UNCDF Cambodia Final Evaluation                                                          47


but are published only on entire projects, so are only marginally useful. For example, Social
Fund primary schools built in Battambang and Pursat averaged $21,400 each, whereas LDF
primary schools in the same provinces have averaged $2,000 to $3,000 per room for new
construction of two to three room schools. Since rural primary schools are roughly the same size
everywhere, size may not explain all of the difference. And although the Social Fund Schools
may well be better built, standards are unlikely to account for order-of-magnitude differences in
cost. The conclusion is that competitive bidding is having a positive effect.


       5.2.3. Procurement

Two important innovations have been introduced to public-sector contracting through the LDF
project. They are pre-qualification of contractors and competitive bidding for contracts.

Pre-qualification has been piloted by Siem Reap province. Contractors apply to the ExCom to be
listed, based on their documented experience, tax payments, legal status, equipment inventory,
history, and references from past work. They are then graded by category (roads, buildings, etc)
and by maximum size of contract (0-5,000, 5-10,000, $10,000 to sky) for which they are allowed
to bid. There has not yet been a case of any legitimate contractor being rejected, though two or
three government staff (who are prohibited from engaging in public works) have been rejected.
There are currently 33 contractors on the list.

The list was originally intended to simplify and make transparent the process of contractor
selection. There had previously been a system of large pre-bid deposits, advance payments, and
retention payments; plus a practice of project owners eliminating contractors based on rumor and
side payments but stated as “experience”. Staff and project owners in the province say it has
achieved its aims.

The list of pre-qualified contractors should also be very important in enforcing quality. It is
difficult to force someone to re-do poor work but easy to give them demerits and prevent them
from future work. According to project owners, CARERE staff, and contractors, the system
increases both cooperation and quality.

Competitive bidding is even more revolutionary. The bidding meeting that the Evaluation Team
observed in Malai district of Banteay Meanchey province was very well run. Sealed bids were
opened publicly and evaluated using four criteria and a weighting system that had been
previously agreed by the council. However, the evaluation table used there may be too complex,
even in its simplified form, for many officials to apply. In fact, DFT staff led the evaluation in
Malai.

The criteria used in Malai included price, experience, equipment, and project duration. They
were combined using a weighting system that effectively downgraded price to the least important
criterion. Prices were indexed to the lowest bid, resulting in a range of 5.71 to 6 for the six
bidders. All other criteria were ranked, resulting in a range of 1 to 6. Each criterion was then
multiplied by its weight, in this case 70% for price and 10% for others. As a result, the final
scores for price ranged from 399 to 420, instead of 70 to 420 as they would have if ranked. The
UNCDF Cambodia Final Evaluation                                                             48


total score for the winning contractor was 559. In this system, even large differences in price
were easily outweighed by a difference of 2 ranks on other criteria. The lowest bidder did not
win the contract.

Furthermore, villagers and commune officials often do not have the information required to
evaluate a contractor’s experience. In Siem Reap, only price is used to evaluate bids and all
other aspects are considered in either the contractor’s qualification process or assumed part of the
contract. For example, it does not matter if a contractor actually owns a cement mixer, since one
can be rented. Therefore, equipment inventories are irrelevant to the bid as long as construction
supervision practices ensure use of the right equipment on the job. This simplification should be
effective as long as contractors are pre-qualified.

Pre-qualification has the additional advantage of providing leverage over contractors to do good-
quality work on time. Siem Reap currently has 28 contractors on their list for various categories
of work. No one has ever been refused a listing, though some have been removed after not
meeting the minimum conditions and others may be removed for unacceptable practices such as
price fixing.

In addition to capacity building for technical staff and government institutions, the CARERE
staff has also been building capacity among contractors. The infrastructure staff of the PSO has
used a proactive approach and holds meetings to ensure contractors are familiar with bid
requirements. They also devote considerable effort to ensuring contractors and TSS appreciate
the requirements of the technical designs and specifications.

Problems with bidding may arise when communes or projects are small. Some communes report
difficulty attracting bidders when there is little private enterprise in their areas and insufficient
incentive to attract bidders from further away.

There is currently a weak connection between work progress and payments to contractors.
Payments should be tied to model schedules and milestones that could be included in the design
templates and standard specifications. Although payments are currently linked to the inspection
program and evaluation reports, the relatively low capacity of the TSS and their high workload
mean that the inspections are either cursory, or become a bottleneck that constrains timely
payments.


       5.2.4. Technical quality

The technical quality of projects varies but in general is good. The CARERE infrastructure staff
in Battambang and Banteay Meanchey have checked over 90% of projects completed as of mid-
1999. Overall, more than 80% of projects were judged acceptable, 18% have problems but can
be fixed, and only 2% are unacceptable.
UNCDF Cambodia Final Evaluation                                                            49



                                               Table 3

                                    Project Quality Assessment

                            Grade                  Battambang       Banteay Meanchey
             Good quality, meets standards             45%                25%
             Small problems but OK                     43%                56%
             Poor quality, can be improved             12%                17%
             Unacceptable.                             1%                  2%
             Projects checked                          93%                96%

The technical quality of projects varies by type. Labor-based roads have not been very good.
This method of construction is not a good option unless VDCs have prior experience. Concrete
work is variable. Building work is generally of a good standard.

Reasons for variation in technical quality include the implementing organization, level of
supervision, and previous experience. Contractors have typically built better-quality projects
than have VDCs, though this depends greatly on the competence of individual contractors.
When VDCs have built their own projects, the quality has often been poor. This has been
especially true of labor-based road construction. The exception is roads built by VDCs by labor-
based methods in areas where people have previous experience with ILO road projects that use
the same methods. Construction supervision should be a daily activity when local capacity is
low, but involvement by Technical Support Staff is necessarily limited. Each TSS member may
cover 10 villages and thus be able to visit a project under construction only two or three times.
Another reason for poor quality in the past may have been a conflict of incentives when VDCs
are both contractor and client. This situation rarely arises now, as VDCs are no longer allowed
to implement their own projects.

The technical support staff does not usually carry out the required post-completion evaluations.
TSS are very busy, and management does not ensure they get done, so evaluations are lower
priority. TSS should enter the evaluations into the LDF database, and CARERE has developed
spreadsheets to help them do this. In practice, only the TSS team leader is able to do it.

       5.2.5. Operation and Maintenance

Maintenance of projects is a perennial problem for all rural infrastructure, and this is especially
true in Cambodia. The Evaluation Team visited many projects that were initially constructed to
an acceptable standard, but were deteriorating after only one or two years without adequate
maintenance. Roads become impassable after a couple of rainy seasons, fences around ponds
fall down, and water pumps break.

Ensuring that maintenance is carried out requires clarity about the roles of project owners and
local government (communes and villages) after the completion of infrastructure projects.
Before new infrastructure projects are started, VDCs and CDCs are required to present
maintenance plans (including financing) to facilitators and support staff and to form maintenance
committees. Follow up is required to make sure these committees do their job.
UNCDF Cambodia Final Evaluation                                                                    50


          Maintenance can indeed be successfully organized locally. A road that totals 9.3 km in
          Phnom Sampou commune of Battambang province was built under the SEILA program
          for sectoral support using ILO labor-based methods. The people who live along the route
          now have plans to maintain it themselves, even in the long empty stretches between
          villages. VDCs arrange maintenance using unpaid workdays. This area is a prime
          orange growing region and people value the road for market access. So they have set
          up checkpoints that allow them to collect tolls from trucks using the road, and to close the
          road during the rainy season when needed. Materials for repairs are bought using tolls,
          and the provincial authorities have also helped by soliciting contributions from the big
          trucking companies that are active in the area.

The Evaluation Team heard little evidence that these committees were active, though some
discussed plans to begin raising funds. Some maintenance funds are collected on an ad-hoc basis
in some areas but are used to respond to immediate problems. Enforcement of collections,
especially for transient road users, would be a big help.

On general principles, operations and maintenance can be funded from either general revenue or
from user fees. User fees are appropriate when users and their activities are readily identified.
For example, trucks are relatively scarce and impose heavy usage, so can be charged road tolls.
But ring wells benefit a large number of people frequently, and are more appropriate for general
fund raising.


          5.2.6. Project Management

There is a need for careful oversight and clear guidelines from line departments, to be used in
both feasibility studies and integration workshops. For example, there should be no short road
segments that don’t connect to a network. But for most project owners, the first they hear of
these is at the integration workshop. If their project is rejected, there may not be time to prepare
an alternative. The Executive Committee should ensure project oversight at the approval stage.

The standard designs for projects have widespread support. That said, more templates with
greater variety are needed to suit local material availability, soil conditions, and hydraulic
conditions. Roads and schools are standard enough and so have few problems, but irrigation
structures vary a great deal. Although there already exists a very impressive set of templates, in
terms of both quality and variety, there is a need for more.

The need arises mainly because technicians don’t know how to vary templates and re-work the
calculations. Only 2 or 3 really understand them and some of the rest even have a hard time
reading plans. So they either apply them as is, which may not be the right thing to do if they
should be adjusted, or make mistakes when adjusting. This implies that further training is
needed. It is clear, however, that no amount of on-the-job training will turn them into fully
qualified design engineers.

Creation of project management committees is required as a means to ensure better quality
control through supervision and certification for payment.13 Some project owners at the village

13
     Biddulph 1999c
UNCDF Cambodia Final Evaluation                                                                     51


level are capable of assisting with supervision. For example, people can be easily trained to
count bags of cement or truckloads of laterite. They should be helped to take on more of this
responsibility.

Project payments to contractors are currently made after transfers from the provincial level to the
CDCs. In some provinces, there is no linkage to actual work progress, whereas in others there
are agreed schedules. The majority of technical support staff has recommended that payments to
contractors be made conditional on satisfactory progress reports.14 In Siem Reap, payments are
made at 50%, 80%, and 100% of completion. This requires enforcement of the system of
inspections and evaluation reports and works well when it is applied consistently.


       5.3. Decentralized Financial Procedures

Under the new SEILA financial system, all project implementation contracts will be through the
PRDC Executive Committee, which will be responsible for management of all investment funds
in the province. This system makes use of the intergovernmental transfer system to manage a
Decentralized Development Fund for sectoral support and the Local Development Fund at the
commune and village level. This is illustrated in Figure 2.

The most significant change from the previous financial system is that PRDC Executive
Committees will be responsible for management of all investment funds in the provinces,
whether they originate from the national budget or from external sources. They will also have
significant influence over line department activities through their financial oversight.

The biggest advantage of the new system is that provinces and CDCs can pay contractors
directly through their bank accounts, or in cash if they wish. The funds stay under regularized
financial management procedures that are subject to internal and external controls. There are
effectively 4 levels of control, including the Ministry of Economics and Finance auditors, an
internal auditor appointed by Governors, external auditors that may be appointed by CARERE,
and the CARERE PSO financial units.

The disadvantage of the new system is that most VDCs and many CDCs may not have the
capacity to manage these financial systems without assistance. Since the system is only just
being implemented in mid-1999, there will be an ongoing need for financial training at all levels.
A review of financial systems may also be necessary at the beginning of every project.

Apparently because of the limited capacity at lower levels, provinces have decided to pay
contractors directly, after certification of progress and completion from the CDCs. This solves a
problem for CDCs caused by the complete absence of banks in two of the provinces. However,
it is not clear if it reduces or increases the risk of LDF funds being misused. There have been
cases of various actors withholding a percentage of payments when they are made in cash. Such
activity becomes impossible if cash is not used.



14
     Crossland, A. “2nd Local Planning Process Review”, Banteay Meanchey Provincial Support Office, August 1998
UNCDF Cambodia Final Evaluation                                                              52


                                              Figure 2

                                      SEILA Financial System




However, there is also a positive argument to be made in favor of project owners making sure

                 RGC Budget (R)                                 External Contributions ($)



                                  Decentralized Development Fund



             National Treasury                MEF                           CARERE
            Provincial Treasury          Special Account



             Provincial Budget



                                    PRDC Executive Committee

                         Cash (R)                               Bank Account ($)



              CDC (cash)                                                    CDC (check)

                                                  Line Depts.

               VDC (cash)                                                  VDC (cash)



                                            Contractors



that the full amount of cash gets to the contractor, as they may feel they have no control or
accountability if payments are made on their behalf.

In either case, there is no way to prevent kickbacks being requested. But CARERE staff
emphasized that there is an important difference between officials being able to hold back a
percentage, possibly on some pretext, versus asking a percentage to be returned to them. The
former is certainly easier and possibly more in keeping with traditional practices.

The key to resolving this dilemma is the process of payment certification. There remains work
to do to convince communes, and some technical staff, to accept certification rather than
payment as the key aspect of their control over project completion and quality. Many officials
remain apprehensive about signing official forms regarding payments, for fear of errors or blame
UNCDF Cambodia Final Evaluation                                                             53


assignation. But once it is accepted as an official responsibility, with the attendant checks and
balances from the provincial Finance Unit, certification should become routine.

Although the system is very new, finance officials in the provinces reported that they believe the
new system will be safer and more efficient than previous practices. With disbursement
managed by the PRDCs, there is a chance that direct payments could lead to delays in payments
when either the certifying or approving officers are absent. However, all provinces should have
appointed alternate officers.

The recent external audit of the SEILA financial system concluded that the basic financial
procedures at all levels are simple and effective.15 The audit also found no evidence of
corruption at any level in the LPP. There were several key points noted in the report that have
been addressed by the most recent revision in the SEILA Financial Manual. These include
reinforcement of review/authorization/certification procedures, better links to budgets and plans,
and better accounting.

Policies for internal auditing, and the person responsible, should be clarified in the provinces.
Since the system is only just being implemented in mid-1999, there will be an ongoing need for
financial training at all levels. There is also the problem of no functioning banks in some
provinces and the resulting requirement that the PRDC manage disbursement.



     5.4. Monitoring and Evaluation

The goal of the monitoring and evaluation (M&E) system is to provide information that enables
CARERE and the Royal Government to manage the project, assess achievements, and track
progress against objectives listed in the project logical framework.

After a slow start, mainly caused by difficulties in finding a suitable advisor after the early
departure of the first M&E advisor, an overall M&E framework was formulated. It details the
purpose and objectives of M&E, definitions, basic principles, CARERE/SEILA policy and
research questions, and the M&E tools and systems developed to carry out the tasks. A
dedicated M&E advisor has been working specifically on the LPP/LDF project in Battambang
and Banteay Meanchey since October 1997. Consequently, development of the project
information systems and databases is behind schedule. However, for the past year, the CARERE
M&E Advisor has been working to consolidate the M&E tools into one system. 16

Because SEILA/CARERE is a large and complex program, the current M&E system is
necessarily ambitious and aims to generate both policy and operational lessons during project
implementation. It is designed for several different tasks: monitoring of CARERE work and
performance, monitoring of PRDC work and performance, monitoring of the local planning
process and outputs, and monitoring of LDF projects. The components of the current system are

15
   Price Waterhouse Coopers. CARERE/SEILA Programme Financial Management Audit: Local Planning Process
Local Development Fund. September 1998
16
   CARERE2 Mid-Term Evaluation, July 1998, UNDP
UNCDF Cambodia Final Evaluation                                                                    54


not well connected and the information collected could be better used in the analysis of the
program. The M&E Advisor and provincial staff are however working to consolidate the tools
into one system.

Monitoring and evaluation work plans have been prepared at several different levels within the
overall SEILA/CARERE project. An overall M&E work plan covers the entire
SEILA/CARERE project. Within that framework, work plans identify the “LPP Sector” which
includes quality assessments of LDF projects and observations of the IPF allocation process.
Some related items also turn up in other sectors within the M&E plans, such as LDF database
training under the Management Sector.

A special M&E framework for the LPP has been developed within the overall CARERE work
plan.17 Tools that have been developed specifically for the LPP include the LDF and RDS
databases; capacity assessments; and special studies into the LPP, LDF allocation, resource
mobilization and collaboration, civil society strengthening, financial management; and finally the
annual internal reviews and external evaluations. This framework for the LPP/LDF includes
indicators, means of verification, reporting schedules, and evaluations.

Despite the somewhat fragmented nature of the formal M&E plans, there has been a great deal of
very good monitoring work done.18 For example, internal studies of the IPF allocation process
are thorough, covering both procedures and quantitative analysis of results. From the
perspective of the LDF and LPP, the most important information resources are the LDF projects
database (for outputs and numbers of beneficiaries) and the Training database (for capacity
building). These sources are detailed and comprehensive. Given that M&E originally had a
negative image as a “policing” activity, these are significant achievements.

In addition to regular program and financial monitoring, the experimental nature of the program
requires that the systems being developed are regularly assessed and adjusted. For example, each
year representatives of all levels of the rural development structure (down to the VDCs) review
the local planning process, identify training needs, and make recommendations for changes. As
a result of regular monitoring of the LPP, the Banteay Meanchey PSO has recommended that it
be reduced from 21 to 15 steps. This revised process is being tested in the 1999 cycle for new
target communes in the province. The reconciliation program in Siem Reap is testing a similarly
reduced LPP.

Impact assessment is the weakest area within current M&E activities, mainly as data for impact
assessment are simply not available. The Commune Database provides a snapshot of one point
in time, but is not yet updated regularly. Therefore, evaluation of socio-economic impact is
impossible from this source, but it does provide a baseline for eventual comparison. There are
other sources of sample data in Cambodia, mainly produced by for specific purposes, which
provide useful baseline data but with limited geographic coverage and smaller samples. 19


17
   Pietersen, I., Local Planning Process Monitoring and Evaluation Framework Proposal, October 1998
18
   Biddulph 1999a, Biddulph 1999b, Pongquan 1998, M&E and LCB Staff 1998
19
   Cambodia Socio-Economic Survey 1997, National Institute of Statistics; Cambodian 1998 Baseline Survey of
CASD Project and WFP Target Areas, UNICEF-WFP; Protracted Emergency Targets Survey 1998, WFP;
UNCDF Cambodia Final Evaluation                                                                55


However, no other source of data is disaggregated to the village level with a broad enough
coverage to be useful for impact assessment of the LDF project.

Monitoring of specific output sectors is even more difficult. There is no complete inventory of
roads or any other category of rural infrastructure in Cambodia. It is therefore impossible to
compare new construction under the LDF to the facilities that existed before the project, nor is it
possible to make comparisons between target and non-target areas. Nor is there unit cost data
(e.g. dollars per kilometer of earth road) available from ministries or other development
organizations for comparison.

In Siem Reap province, a new experimental procedure for the formation of CDCs has been
proposed that will include village representatives. These representatives will verify and update
data in the village database. The DFT/PFT will facilitate and train the villagers in this exercise.
This could be the beginning of an ongoing localized M&E capability.

At this stage, most of the monitoring and the production of reports is managed by CARERE PSO
staff, though there are plans to transfer responsibility for M&E to the PRDC and active efforts in
several areas. CARERE staff have been working closely with provincial departments of
planning to develop and implement monitoring of the Provincial Development Plans, SEILA
Investment Plans, and integration workshops. The provincial departments of planning have also
taken over responsibility for the Commune Database, and villages and communes have
responsibility for annual collection of data and monitoring of their projects. Contact persons in
the RDS include an M&E Assistant at each PSO, focal persons at each provincial department,
and an M&E unit within the ExCom Secretariat.

The ITAD Evaluation Report commented that M&E systems are too complex for a local
development project. Certainly the reliance on CARERE staff cannot continue if provincial
departments of planning are to take over data collection and analysis. But for such a complex
policy experiment this complexity was probably warranted.




Household Survey of Management of Freshwater Capture Fisheries of Cambodia 1995-96, Mekong River
Commission
UNCDF Cambodia Final Evaluation                                                             56




6.     EVALUATION OF IMPACT: OVERALL PROJECT RELEVANCE
        AND EFFECTIVENESS


     6.1. Institutions

        6.1.1. Provincial Level Institutions


     6.2. The local planning process

The LPP has always been open to other “financial windows” besides SEILA. There is now
considerable interest in using the LPP for planning and coordination of multiple sources of
investment in villages and communes, through a Decentralized Development Fund. This is
supported by recent decisions by the Royal Government, IFAD, and World Bank to capitalize the
LDF for new projects. Since there now exists an intergovernmental transfer system to the
PRDCs, through the Treasury, separate from CARERE/UNDP/LDF as part of the SEILA
structure, the LPP could become a universal process for selecting priorities from any source.

The local planning process can also eventually form the basis for integrated planning at all levels
within provinces. There are already resource allocations being made more often at the commune
level than at the village level, and the integration workshops are considering sectoral and outside
support as well as LDF funds in their decision making. Eventually this process may form the
basis for integrated provincial development plans.

The experience to date has been that the LPP process has been modified in each of the SEILA
provinces to suit the local context. In the next phase of the project, it will be important to begin
developing a general framework for national guidelines that still allow local flexibility.

        6.2.1. Factors of Success

        •   To ensure participation, dialogue and accountability, the LPP is necessarily complex.
            The establishment of clear guidelines for procedures that have been drawn up for the
            LPP are important factors in giving VDC and CDC members an objective basis for
            decision making and a justification that can be made clear to everyone of why
            projects are selected. While these guidelines need to be improved, the Evaluation
            Team recognizes that much of the success of the LPP is due to the efforts of the
            SEILA staff and the practice of continuous review and improvement of the LPP

        •   The availability of resources to government departments through the SEILA
            Investment Program (SIP) plays an important role in the success of the LPP even
            though it falls outside the LDF. Without the SIP, the program would be out of balance
            and the success of the District Integration Workshop would be greatly limited.
UNCDF Cambodia Final Evaluation                                                            57


       •   The introduction of the District Integration Workshop, with the participation of
           CDCs, departments and NGOs, greatly enhances the ability of the LPP to foster an
           integrated approach to the development of the commune.

       6.2.2. Transition Strategies for Mature Communes

One of the questions the program has to deal with is to what extent the support of the facilitators
can be reduced. A number of factors come into play around this issue that require consideration
in elaborating a long-term strategy for support to the communes.

First, if the program is required to increase its coverage additional facilitation support will be
needed. Adding more facilitators will only be possible if more resources become available. The
capacity of the facilitators is already stretched as it is.

Second is the issue of how new elections will deplete the skills of the VDCs. At the time of this
writing, re-elections have only taken place in the provinces of Battambang and Pursat. While all
the data is not in yet, it appears that in a number of villages all the VDCs members chose not to
stand for re-election, usually because of other domestic demands on their time. If this turns out
to be a trend, it will have important implications for the need for facilitation. If the committees
lose their institutional memory, facilitators will find themselves training committees from ground
zero. The option of staggering the election of members was discussed, but the facilitators said
that staging elections more often than every three years would also make demands on their time.

This subject merits further consideration. Possible solutions include a period of overlap between
newly-elected and outgoing members to facilitate transfer of institutional memory, activities that
invite occasional participation of former members, or providing facilitation skills to one or more
villagers. Finally, the shift to the commune focus and the creation of the Commune Councils
will undoubtedly effect changes in the way facilitation is managed.

Given these considerations, at this time, the Evaluation Team is not in a position to make specific
recommendations regarding a strategy for reducing support to the communes. However, as more
information becomes available over the next year it will be important for the program to
formulate a long-term strategy for support to the communes as the question is bound to come up
regularly.


       6.2.3. Resource Allocation

The use of objective criteria for poverty and population for IPF allocation is becoming
established but needs further improvement. CDCs need ongoing training in use of data and
methods.

Allocations at both commune and village level should be weighted by actual population, and not
simply the number of villages. Poverty criteria need refinement and probably simplification, but
further statistical analysis is required to make a recommendation.
UNCDF Cambodia Final Evaluation                                                            58


A weighted average of the poverty ranking and population is currently used at the provincial
level in two provinces but not in three provinces. Since only a few criteria are used in practice at
the commune level, and capacity for more sophisticated methods is low in most places, a small
number of simple criteria could be adopted as the standard. Consumption-related variables such
as housing condition, motorbikes, or household equipment are equally good as indicators of
relative wealth and poverty.

A different way to assess village poverty is to use a matrix ranking as part of the participatory
rural appraisal process. In this approach, villagers are asked to rank the poverty of all other
villages in their commune in a public meeting. This has been used successfully in Vietnam and
seems to generate widespread public understanding and acceptance of the outcome.

Infrastructure indicators should be added to the criteria for allocations. Current poverty
indicators such as the number of children attending school may actually reflect transportation
problems instead. An inventory of classrooms relative to population, all-weather roads, and safe
drinking water sources would be sufficient, and would have the advantage of being a normal
responsibility of line departments and regularly updated at the start of the annual LPP. Since the
LDF is intended to fund rural infrastructure, allocations should consider the existing state of
similar public facilities.

The IFAD evaluation notes a potential conflict between a bottom-up village planning
methodology and planned poverty targeting by higher levels of government. This conflict has
been ignored in the allocation process that attempts to target poor communes. It will always be
the case that there are well-off families in poor communes and poor families in better-off
communes, so this conflict will always exist in any methodology that relies on aggregate data.
Unless planning and project interventions can be targeted at individual families, it is
unresolvable

       6.2.4. Monitoring and Evaluation

M&E work plans now identify separately the outputs of the LDF process to allow impact
analysis of this project. Given the current importance of the LPP in the overall CARERE
program and the fact that it was derived from the Local Development Fund approach, it is
important to assess the contribution made by participatory methods. Items recommended for
monitoring should include a range of UNCDF-recommended key performance indicators (KPIs)
for good governance and institutionalization of planning.

While the evaluation team is impressed with the activities observed in the field, observation and
qualitative reports are not a substitute for concrete information when it comes to measuring
impact. Although an impressive array of information was collected at the grassroots level at the
establishment of every VDC, a second round of data collection has yet to be conducted in most
areas, limiting the Evaluation Team’s ability to discuss impact.

Data for impact analysis must be collected and regularly updated as part of the annual planning
process. Since there is an annual review of development plans at the village, commune, and
province levels, data from the Commune Database could also be updated at this time. The
UNCDF Cambodia Final Evaluation                                                           59


simplified subset of questions recommended for reconciliation programs in Siem Reap is a move
in the right direction. The results will most likely reinforce the positive impressions of the
various evaluation teams and add legitimacy to the project.

Key performance indicators that relate infrastructure provision to national standards should
become part of the LDF database. As a complementary activity, TSS must be required, and
given realistic opportunities, to carry out the post-completion reports.


       6.2.5. Considerations for the Future

SEILA staff, both government and CARERE, play many roles within the LPP: innovation,
implementation, training, establishment of systems, and documentation. The SEILA staff has
performed well under pressure. Now faced with a situation of dwindling resources and limited
technical assistance, they are being asked how they can move to scale.

An important aspect of consolidation is in-house capacity to build skills. Each province has on
average 50 to 60 provincial and district facilitators. It is not clear that CARERE is focusing on
building capacity at the provincial level to train new facilitators as the program expands or as
facilitators are replaced. It will not be possible to further reduce the need for CARERE LCB staff
until this capacity is established.



   6.3. Delivery of rural infrastructure

       6.3.1. Project Impact

The projects that are being completed have passed their first and possibly most important test.
They are widely believed by beneficiaries to meet their needs and improve their situations.
Roads have opened access to isolated areas, making possible better marketing of local products
and improving prices for local people. Roads and bridges are also enabling more children to go
to school regularly. Irrigation projects have raised yields and in some cases allowed a second
annual crop. As a result of perceived positive impacts, people have been making their local cash
contributions regularly and in some cases raising them every year.

This evaluation of the overall project should include quantitative evidence to support these
impressions of the impacts of infrastructure projects. Field interviews and qualitative statements
in internal reports certainly tend to support them. As discussed above, however, the information
systems and existing data do not yet permit such impact assessment. Now that a baseline exist,
ongoing monitoring make such assessment possible in the future.
UNCDF Cambodia Final Evaluation                                                                 60


       In Siem Reap province, villagers in Dan Run commune province have combined their
       LDF allocations to build irrigation structures along the ‘78 Dam, a Pol Pot era dike several
       kilometers long. Their projects have been mostly culverts with gates and a larger
       bridge/gate. Water is stored for dry-season rice, which they could not otherwise grow.
       This has created a crop where none existed before. The new culverts and gates have
       improved the farmers’ ability to regulate water. Five villages benefit from this scheme, so
       it is in effect a commune project, and a water user group operates the gates. The group
       taxes members for operation and maintenance, and has included the new structures in
       their maintenance plan.



The Evaluation Team believes an evaluation of the overall cost effectiveness of the LDF project
to be an impossible task within the constraints of the Evaluation Team. It is simply not possible
to separate the specific activities of the LPP from the wider CARERE project. Staff time, capital
equipment, and operations are all intertwined and would require both financial and management
audits to disentangle. The IFAD Evaluation report commented that project overheads in the
original Project Document were 50 per cent, plus these other costs. If attempted, it is possible
that rough estimates of expenditure per project or per impact could be produced, but it is not
clear to what they might be compared.


       6.3.2. Project types

Non-infrastructure projects should be curtailed. The risk of them becoming private property is
real, and the damage that such an outcome would do to the legitimacy of the process is
significant. The LPP is not well suited to managing these types of projects. Furthermore, it is
probably not realistic to expect that these types of projects would be continued by the public
sector when the LDF approach is replicated to cover all provinces. The goal of the LDF is to
develop a sustainable structure for decentralized planning, not to deliver short-term economic
development.

There is a need for careful oversight and clear guidelines from sectors, to be used in both
feasibility studies and integration workshops. Project oversight at the approval stage varies. It
should be done by the ExCom members.


       6.3.3. Procurement and financial management

Bidding procedures should be simplified. Existing criteria should be used to screen and pre-
qualify contractors. Contractors should be pre-qualified according to the category and size of
work they are allowed to undertake. Bids should then be evaluated on price only.

Contract payments should continue to be made directly from the PRDC ExCom to contractors,
following a certification process, in order to facilitate internal and external audits. Payments
should be tied to model schedules and milestones that could be included in the templates.

The new financial control system is straightforward and safe. It does, however, place
considerable responsibility in provincial Financial Units. There is a chance that a lack of
UNCDF Cambodia Final Evaluation                                                              61


financial and accounting experience in those offices could lead to bottlenecks in contract
management. There would seem to be an urgent need for training in the new system for all those
involved, a fact reinforced by all the provincial Finance Directors interviewed.

There remains some work to do to convince communes, and some technical staff, to accept
certification as the key aspect of their control over project completion and quality. Written
guidelines should be developed that clearly set out responsibilities for each link in the chain, and
remove as much uncertainty as possible regarding the consequences of errors. In addition,
policies for internal auditing, and the persons responsible, should be clarified in the province.
Governors must begin to appoint internal auditors.


       6.3.4. Project Management

Improvements in construction supervision staff and practices are needed to raise overall quality
and shorten project schedules. This must be done by the TSS, but their territories are difficult to
cover with adequate frequency. As it is difficult in some provinces for technical support staff to
cover their territory, it may be necessary to base most of them at district centers.

There is a need for more templates with greater variety, especially for irrigation, to suit local
material availability, soil conditions, and hydraulic conditions.

Weekly review meetings and special training for technical staff are both appreciated by those
who have been involved. More emphasis should be placed on this assistance role for both TSS
and contractors.

Creation of project management committees, possibly as sub-committees of the VDC, should be
reinforced. They are responsible for some supervision and certification for payment and need
technical support to increase their effectiveness.

Maintenance of projects is a perennial problem. Maintenance plans and revolving funds don’t
exist but should. Enforcement of collections, especially for transient road users, would be a big
help.

Scaling up to cover all communes in a province and to cover reconciliation areas will require
increases in operational complexity, but it is not clear whether the increases will be proportional
with the number of additional communes. In Banteay Meanchey, new communes will not be
further away than are existing communes, so the problem is simply one of additional staff days.
In the other provinces, additional communes will be much further away from the provincial
centers than those currently in the program. In some cases, travel times will increase greatly, and
remote visits will require overnight and multiple day stays. Getting technical staff out to remote
locations cannot be done in a day if they are based at the provincial centers, so they should be
based at district centers more often.

Continued reliance on contractors will be essential for mainstreaming and upscaling of the LDF
concept. It shifts the burden of delivery onto the private sector and at the same time gives
UNCDF Cambodia Final Evaluation                                                         62


incentives for the development of technical capacity. However, it also raises the burden on TSS
to be better project managers. Continued training in management as well as technical designs
will be necessary.



       6.3.5. Factors of Success

       •   The PRDC has become the executive authority for all investment funds in the
           province, with implementation taking place though the CDCs and VDCs in the same
           way for all outside resources.

       •   The policy to require expanded use of contractors has led to much more involvement
           of the private sector in public works and development of the sector in general.

       •   Competitive bidding is saving money for project owners. The average cost savings to
           date are modest, but individual projects have seen savings of 20 per cent and more.
           Anecdotally, several communes are reported to have saved enough money on
           approved projects to fund construction of additional minor projects.

       •   Proactive technical assistance on the part of the CARERE infrastructure staff is
           having significant positive effects on construction quality and bid outcomes.
UNCDF Cambodia Final Evaluation                                                             63




7.     NEW DEVELOPMENTS

     7.1. Reconciliation Areas

One of the most telling achievements of the SEILA program is the way it has responded to the
challenge of inclusion of the former Khmer Rouge areas. What is remarkable is the rapid trust
that was built up between former enemies, provincial staff, and CARERE management and staff,
as well as the rapidity in which the VDCs and CDCs were established.

It is also remarkable how quickly the people in the Reconciliation Areas “took” to the
participatory process of SEILA. People in the VDCs, CDCs, as well as the District Governors
and Deputy District Governors demonstrated evidence of their enthusiasm. In one CDC meeting
a woman said, “In the old days you were told what to do and how to do it.” (As she said this, she
stood snapped to attention, saluted and said) “Yes, Sir.” She then sat down and continued, “Now
we sit around and discuss, listen to each other and share ideas. It is a totally different situation.
We all feel like we are part of it.”



     7.2. The New Sub-Decree and SEILA/CARERE

A Sub-Decree (No. 78ANKR.BK) signed by the Prime Minister, Hun Sen, on August 23, 1999,
reaffirms the previous Sub-Decree. It also allows for the following:

•    Confirms that the SEILA program sits atop the Rural Development structure and that the
     PRDC reports to the STF.

•    Allows for the expansion of the project into other provinces subject to availability of funds.

•    Establishes the PRDC ExCom and confirms its role as the body that manages implementation
     of the SEILA program at province level.

•    Establishes the DDF under the Ministry of Finance; defines the role of the STF Secretariat in
     relation to the DDF as assisting the MoF in mobilization of resources, reporting to donors,
     monitoring and auditing.


     7.3. The New Commune Councils beginning in the year 2000

Laws are being drawn up now by the RGC for the establishment of over 1,600 popularly elected
Commune councils (CC) in 2000. Rather than being a committee like the CDC that oversees
development, the CCs will be the official administrative units of government with powers to tax
and collect other revenues, maintain their own budget, provide a police force and carry out other
official local government functions. Although the law is not passed nor have its contents been
made public, the general consensus is that it will have some of the following:
UNCDF Cambodia Final Evaluation                                                          64




• A Chairman who will be elected by the people. (The person who receives the most number
  of votes will become Chairman).

•   A Deputy Chairman will be elected by the council.

•   A Commune Clerk will be appointed by national government.

•   It will have between 9-11 members with 50 % women.

They will come under the umbrella of a national inter-ministerial committee headed by the
Ministry of Interior. The existing VDCs will likely be a committee within the CC representing
development for the commune. It is also likely that all development will still be coordinated
through the PRDC as in the SEILA/CARERE/LDF provinces.

The prospect of 1,600 CCs coming into being all at the same time raises an array of significant
issues among which are:

•   Will this cause the VDCs to lose some of their identity and importance if they are simply a
    committee within the CC?

•   Will the VDCs still be elected by popular vote?

•   The vast majority of the new CCs will have little training and technical support that was
    provided for in the CARERE provinces. Most members will have little idea of their roles
    and responsibilities, much less the capacity of managing a unit of government with all its
    ramifications.

•   This raises the big question as to how much the SEILA/CARERE/LDF project can assist the
    government in training, technical support, TOT, and other services for the new CCs.

Some of the issues will be raised again in the section on Future Policy Formulation.


    7.4. The Experimental Commune

There has been much discussion about the amount of time and effort (usually about ten days) that
goes into setting up a VDC, collecting data, developing village projects, etc. Also, once the
VDC is established, so the critique continues, a DFT is usually expected to conduct all of the
subsequent meetings, taking his/her time from other important duties. Among other VDC issues
provided in this report are that VDCs don’t fully represent the village, and the only reason to
meet is to formulate village projects.

The Experimental Communes being developed in Siem Reap attempt to answer these issues by
eliminating the VDC altogether, by putting more focus on the communes and reducing it on the
villages. One man and one woman are elected by each village as representatives of the village to
UNCDF Cambodia Final Evaluation                                                         65


the CDC. The village is represented, but the long process of setting up the VDC is eliminated;
and all training is done at the CDC level, thus eliminating much time and expense. This also is a
possible strategy for getting the 1,600 new Commune Councils to be elected in 2000 on board as
quickly as possible.
UNCDF Cambodia Final Evaluation                                                              66




8.     OVERALL CRITICAL AND STRATEGIC ISSUES

     8.1. Sustainability

There are several key elements that provide a great deal of sustainability to the project:

♦ A new decentralized rural development structure is effectively providing, on the whole, good
     quality infrastructure throughout its target areas.

♦ Provincial administrations, through the PRDC and its other bodies, are continuing to build
     the capacity necessary to continue this process. National staff in both CARERE and
     government are increasingly taking their own initiatives and taking over more of the
     management of the projects. As this process continues, CARERE can reduce its technical
     support, training, and catalytic involvement, and take more of an advisory role.

♦ Already, two of the five Provincial Program Managers resigned providing the positions to
     Cambodians. Next year, the PPM in Siem Reap will step down to be replaced by a
     Cambodian.

♦ Managing rural development is increasingly being taken over by the ExCom; and technical
     support and facilitation is already provided by the Secretariat.

♦ CDCs and VDCs are demonstrating the ability to make plans for their communities in a
     participatory and transparent manner.

♦ Beneficiaries are consistently making adequate local contribution to all projects.

♦ The Royal Government of Cambodia has officially recognized the RDS structure, which is
     directed by SEILA, as the official structure in the five provinces in which SEILA operates.

♦ The Royal Government of Cambodia has funded over $200,000 in 1999 and is considering
     $400,000 to $500,000 in the year 2000.

♦ All local projects implemented by the various ministries in Ratanakiri will require a 10%
     local contribution, and in Battambang all road projects will require a 20% local contribution.

If, on the other hand, the SEILA/CARERE/LDF project is simply going to be one stand- alone
project that eventually needs to be phased out, different strategies need to be addressed. Much of
this again is dependent on how quickly the RGC moves on the administration reform and
demobilization programs. This depends on the government’s political will and a number of other
grave issues that will not be addressed in this report.
UNCDF Cambodia Final Evaluation                                                             67


       8.1.1. Administrative Reform and Demobilization

The ultimate answer to sustainability lies in the government’s ability to seriously begin its
process of administrative reform and demobilization. As described above, the government
payroll is totally out of synch with the actual needs of government to perform its duties. The
army estimated 145,00 soldiers on the payroll but in reality has only an estimated 90,000 actual
soldiers, the rest being “ghost soldiers” with certain generals getting the extra salaries. Some
unofficial reports say that the government will reduce this number by 55,000 over the next three
to five years beginning in the year 2000. This could bring in close to US $800,000 yearly into
the government treasury. The government has been talking about an administrative reform
program for many years; but with the volatile political problems plaguing the process, virtually
nothing has been accomplished. The administrative reform program would reduce the
government payroll over a period of time in the same manner as the decentralization plan
described above. With some degree of political stability in government now evident, there could
be a real chance for this process, along with demobilization, to begin.

If and when these reforms begin to take place, thousands of people will be taken off the payroll.
Government staff performing full-time work with clear roles and responsibilities could have their
salaries increased. These salaries could be incrementally increased as other staff, with little or no
responsibilities, are dropped from the payroll. As this process occurs, salary supplements in the
SEILA/CARERE/LDF could be proportionally decreased.

In addition, as government staff matures under the support of CARERE, (a situation that is now
occurring), CARERE can begin to reduce its staff or use it to move into other areas that the
government wishes to open up to the SEILA/CARERE/LDF process.


       8.1.2. Generating Income from Local Taxes

The new laws being drafted will provide for the right of the Commune Councils to collect taxes.
It is understood that they will focus on property taxes in provincial and district towns. It is also
possible to collect taxes on market vendors and even income of wealthy merchants and farmers.
This income could provide needed funds for infrastructure development that is now being
provided by donors.


   8.2. Expansion of SEILA/CARERE/LDF Model into Other Provinces

The following questions, among others, arise when considering the future of
SEILA/CARERE/LDF:

1. Will RGC want to move this SEILA/CARERE/LDF process into the other seventeen
   provinces?

2. If so, would they use the same modality that they used in the five SEILA provinces with
   CARERE as the trainer, technical advisor and catalyst?
UNCDF Cambodia Final Evaluation                                                           68




3. If the answer to number three above is “yes,” many questions arise as to how this will be
   implemented.

4. Do they want CARERE’s assistance to help strengthen the more than 1,460 new Commune
   Councils (that will not have had CARERE’s assistance) that will be established next year?

If the answer to any of the above questions is positive, strategies must be worked out between the
RGC and CARERE along with other donors to implement this vision.


   8.3. Transition Strategies

LDF funding is due to end this year while the SEILA/CARERE project is funded through the
year 2000. The project needs LDF funding to take it through the year 2000.

Year 2000 should be spent consolidating current activities, bringing the main experimental
aspects of the local planning program to a close (perhaps offering a series of options within the
model). SEILA/CARERE/LDF should decide on how many communes and villages they can
practically bring under the LPP and LDF and complete strategic plans with time frames and
budgets that address this. It is also important that CARERE take measures to avoid losing staff
during the period of uncertainty before another project is signed, namely ensuring that the
formulation of the next phase is completed as early in 2000 as possible.

Emphasis in the year 2,000 should also focus on building training-of-trainers’ capacity within the
facilitation teams at the provincial level, on working with the PRDC to establish how best to
organize the facilitation work load (work load) and supervisory responsibilities, and on
developing a clear strategy that will allow the PFTs and DFTs to reduce their role in mature
communes.

It is quite conceivable that within three to five years, CARERE could cease to exist, as it is
known now. The awkward problem is that money is needed to fund provincial, district,
commune and village programs and projects to provide work for the government staff who have
been receiving this training and support. The institutional structures are in place and are
functioning, and the government staff has the capacity to perform the work, as do the members
of the various committees. Funds need to come from donors and eventually from government,
otherwise, the institutions that have been established will lack one of the key elements that make
their existence meaningful.

Any exit strategies within the next five years will be difficult, particularly if the government
wishes to replicate the process in other provinces. However, if the Administration Reform and
Demobilization Programs begin to be implemented, funds from those salaries can be channeled
into higher salaries for the civil servants that actual work full time. As these funds continue to
increase over time, LDF and other funds can be proportionally reduced. In addition, as stated
above, government staff is increasingly taking over more responsibility. CARERE can continue
to play more of a true consulting role rather than a continuous trainer and support role.
UNCDF Cambodia Final Evaluation                                                             69




This can also be true if SEILA/CARERE is called upon by government to help other provinces
replicate the project. SEILA/CARERE will play a lesser role in their own provinces while
sending key staff into other targeted provinces to assist in the establishment of the SEILA
program.

If donor funds are reduced sharply over the next few years, and little or no money comes from
government, it is likely that the VDCs will not sustain themselves. On the other hand, with the
formation of the Commune Councils as administration units of government; it seems likely that
they will be sustained. This is particularly true if they are granted taxing powers, which is likely.
Government is considering taxing real estate in provinces, districts and even small towns. This,
plus a market tax and even some form of income tax for families with large pieces of fertile land,
draft animals, etc., could generate income for development.
UNCDF Cambodia Final Evaluation                                                              70




9.     RECOMMENDATIONS

•    There is a need to emphasize villages and communes doing more self- help projects
     themselves, but with the technical support of local governments. It should be made clear to
     CDCs and VDCs from the beginning what the government will do and where their
     responsibility ends. Village responsibilities should be clearly stated. The allocation process
     could be altered somewhat to reward villages and communes that initiate and complete
     projects themselves. This could be done during the second and third cycles. This would
     perhaps help people to build their sense of community while at the same time developing
     more of a self-help attitude.

•    More emphasis could be put on villages seeking outside help or doing more work themselves
     on projects where LDF allocations were insufficient to cover total costs. For instance, if the
     allocation for a village was only $4,000, the project cost $7,000, and the village wanted it
     badly enough, they should be encouraged to seek other funds or do more of the work
     themselves.

•    From the beginning, there has been much skepticism and criticism from the NGO community
     about the SEILA/CARERE/LDF project. Many of the NGOs still view the project from the
     standpoint of “community development” rather than a capacity building, decentralization and
     local government. There appears to be a strong need to inform the NGO community of the
     real mission of the project. The ET realizes this is a difficult situation and not easy to
     correct; however, it is hoped that the project would develop a public relations strategy that
     will more fully address this issue.

•    VDCs often do not represent the majority of the people in a village. The members are
     usually elected because they can perhaps read and write a little, and this is thought necessary
     to submit projects to the CDC. The VDC is often not connected with the Wat Committee or
     other groups that may have influence in the village. New strategies should be developed
     whereby more of the total village is represented and where more inclusion of traditional
     leaders is advanced.

Where NGOs have assisted in establishing small committees, there is often the complaint that the
newly established VDCs often ignore these committees. It is recommended that part of the
orientation training deal with this issue, and that a formal mechanism be set up where these
groups are informally part of the committee.
• It should be made clear from the start that after the first few VDC meetings, all meetings
    should be run by the VDC Chairperson and not the government facilitator. Also,
    chairpersons should be trained at the CDC level on how to conduct participatory meetings
    and how to keep different groups involved.

•    Give villages a clear explanation of the CDC and its roles and responsibilities, as well as the
     LDF and how it works. In addition, show the villagers the commune map and explain:
•    How it was created and what it means so they have a clear understanding of the whole
     process.
UNCDF Cambodia Final Evaluation                                                          71




•   Villagers should be notified well in advance of all VDC meetings, and should always be
    involved in decisions on such things as where wells and ponds should be located, how
    contributions should be collected, and how projects will be maintained.

•   The members of the SEILA team should consider spending the year 2,000 consolidating
    current activities, bringing the main experimental aspects of the local planning program to a
    close (perhaps offering a series of options within the model), and thinking about ways to
    work strategically- given existing constraints and demands. It is also important that CARERE
    take measures to avoid losing staff during the period of uncertainty before another project is
    signed.

•   The three provinces visited by the ET demonstrated that they can probably complete the
    expansion of VDCs and CDC within two years. As they move to set up new committees, the
    older committees become more mature and need less attention from government staff (PFTs,
    DFTs and TSSs); therefore, this staff can spend more time assisting new VDCs and CDCs.
    This is also true of CARERE’s local capacity building staff. One important exception is that
    new members of VDCs will need training and assistance. This is also true where, under a
    new election, all of the elected members are new. This suggests the gradual reduction of
    both CARERE and government staff over time. It also means that as government staff
    matures in proficiency, less is needed of CARERE, since they are there primarily as trainers,
    support providers, and advisors.
UNCDF Cambodia Final Evaluation                                                            72




10. FUTURE PROJECT FORMULATION


   10.1.       One Year Full Time Local Government/Decentralization Advisor

UNCDF/UNDP or other donors could provide a full time local government and decentralization
specialist for one year to work between SEILA, CARERE and national government to assist
them in the formulation of plans to take the project into 2,001 and beyond. The role would be
for a technical advisor in local government and decentralization and would require facilitation
skills. The person would live in the country, network between different agencies, serve as a
catalyst, and assist in the development of a three- to- five- year strategic plan to which donors
could respond. Some of the possible issues the person would deal with are:

♦ The experience to date has been that the LPP process has been modified in each of the
   SEILA provinces to suit the local context. In the next phase of the project, it will be
   important to begin developing a general framework for national guidelines that still allows
   local flexibility.

♦ Rather than reduce present SEILA staff, it is possible that staff should be increased to
   provide key people to go into other provinces to train local leaders in the methodology.

♦ Reinforce the new Decentralized Development Fund at the national level that will be a
   depository from an expanded list of donors from which SEILA could work. This would
   include more strategies for making SEILA the vehicle for delivering development through
   local government that many donors could respond to.

♦ SEILA has established a decentralized planning and financial structure that has proven
   effective and is potentially a national model. It is hoped that the model will continue to gain
   support from the Royal Government and will be allowed to adjust to changes in the local
   structures anticipated by the upcoming Commune Elections, all the while maintaining
   responsibility for coordination of development activities at the provincial level. This would
   mean that the elected Commune Councils would coordinate their development activities
   through a non-elected body.

♦ With the establishment of the new Commune Councils, there is a need to identify and
   coordinate outside financial resources for development projects, including multi-sectoral, at
   the commune level.
UNCDF Cambodia Final Evaluation                                                            73




11.    CONCLUSION

SEILA/CARERE/LDF has been an enormously ambitious project. It has also been costly. It has
attempted to almost remake a local government process and deliver vitally needed infrastructure
development at the same. Although there is scope for improvement and much hard work yet to
be done to institutionalize the structures and processes that have been created, the creation of this
system has been a major achievement and has for the most part achieved its ambition. The
SEILA/CARERE/LDF system has accomplished the following:

1. The project has more than achieved all of its goals and objectives and the outputs described
   in the main phase document.

2. It has assisted the governments in the five SEILA provinces in establishing a participatory
   decentralized local government planning and development system in a government/society
   known for its top down hierarchical style of rule.

3. It has implemented a precedent in local government in Cambodia whereby a certain
   percentage of government committee members must be women.

4. The system created has been formally accepted by the RGC and is looked upon by most
   government officials as the model for decentralized local government in all provinces.

5. It has had a major influence in the drafting of new local government laws and with the
   formulation of the Commune Councils in the year 2,000, which is seen as move towards
   democratization of local government.

The major task of SEILA/CARERE/LDF is now to consolidate progress and to assist and work
with the Royal Government of Cambodia, as the government desires, and other donors in the
formulation of strategies and plans to move the SEILA/CARERE/LDF project forward into the
year 2,000 and beyond. In addition, to assist the government, if it so desires, in its move towards
decentralization in other provinces.
UNCDF Cambodia Final Evaluaton                                                             i


EVALUATION SUMMARY

I.     Basic Project Data

       Country          Cambodia      Project Number: CMB 97/Co1

       Project Title:                 Local Development Fund

       Sector                         (0210) Development Strategies, Policies and Planning

       Sub Sector                     Local Governance

       Government Executing
       Agency                         SEILA Task force

       United Nations
       Cooperating Agency             UNOPS

       Financing
              UNCDF                   US$ 4,5 million
              UNDP                    $US 20,5 million
              Others                  $US 35 million (as trust fund and cost sharing
                                      contribution)

       Approval Date:                 April 1997
       Starting Date                  August 1997
       Completion Date                December 31 1999
       Evaluation Date                October 1999




1. BACKGROUND OF PROJECT

For almost thirty years Cambodia has been in a state of flux beginning with the overthrow of the
government in 1970. People have experienced periods of civil war, a radical government
attempting to remake society and culture, famine and genocide and an invasion and occupation
by an outside nation. They later saw the signing of peace accords in Paris between warring
factions and the establishment of a peacekeeping force by the United Nations, followed by hotly
contested national elections and the establishment of a fragile coalition government. The 1996-
97 period resulted in the violent ousting of some members of one party in a coalition
government, and the beginning of the breakup of the former Khmer Rouge, followed by its
partial integration into the nation’s government. With the national elections in 1998 helping to
consolidate the power of the ruling CPP party along with the main opposition party outwardly
willing to take a lesser part in government, there is an appearance of consolidation and stability
in the country. In addition, the reentry of the former Khmer Rouge into Cambodian society has
UNCDF Cambodia Final Evaluaton                                                             ii


brought about a long- desired peace. Warring between politically active groups has subsided,
and the roads are almost free from renegade soldiers and bandits.

Except for the period immediately after the political violence in 1997, the economy, since the
1993 elections, has been growing at the rate of about 6 % per year; and the Riel has been
stabilized. Per capita income is up over $ 280 compared to $130 in 1990. Around 85% of the
population live in the rural areas, and most are subsistence rice farmers. Growth, however, is not
yet broad based. It is concentrated in the capital in construction and services and is largely aid
driven.

These years of war and civil strife have left Cambodia’s infrastructure in shambles, with
secondary roads in constant need of repair and with main arterial roads in need of upgrading. To
travel from one part of a district to another in the rainy season may take as much as two to three
hours. Irrigation structures have been poorly designed and broken down leaving most farmers
depending solely on rainfall for their crops. In addition, crop loss due to drought and flood has
become common. Only a small percentage of the population has access to reliable water for
domestic use. Schools are in a state of disrepair with leaky roofs and flooded classrooms during
the rainy season, and children have problems getting to school because of the poor condition of
roads.

The greatest concern is the effect on the capacity of human resources and institutions, the
economy, society, culture, religion, and the sense of community. Basic trust between people in
villages has even been affected. Most of the educated people were eliminated, died of starvation
or left the country. The same fate happened to over 75% of the teachers. These events have had
an enormous influence on the process of development as well, and seasoned development
workers in Cambodia are adamant in advising that projects often take much longer and require
much more effort than in other developing countries.



2. DESCRIPTION OF THE PROJECT

The project is a continuation of the pilot project in two provinces for the establishment of a Local
Development Fund with an initial allocation by a grant from UNCDF of US$3 million over a
period of three years from 1997 - 1999. Given the significant and positive lessons learned during
the pilot phase and the adoption of the LDF/LPP methodology by CARERE, this amount is
increased by US $1,426,400. This is to allow for a greater amount of infrastructure projects and
to provide for two Technical Assistance Advisors and a Technical Backstopping Monitoring
arrangement with a nearby technical institution.

With two communes in each of the two provinces already receiving LDF funding, the project
provides for an expansion into 8 – 10 additional communes in 1997 and at least that in 1998 and
1999. The grants support the process of decentralized planning under the LPP in the form of
Indicative Planning Figures (IPFs) to the CDCs in the amount US $50,000 for projects that have
characteristics of “public goods.” The funds are allocated over a three-year cycle. To qualify,
CDCs must have complete coverage VDCs or similar village committees sponsored by NGOs.
UNCDF Cambodia Final Evaluaton                                                             iii


The specific allocation to each commune depends on the population size, the number of villages,
and the amount of development which has already taken place. Communes and villages
receiving projects are to make at least a 10 % contribution.


       A. Development Objective

The overall Development Objective is:

To alleviate poverty and spread peace in Cambodia by strengthening the bonds linking society to
the structures of the State and empowering the Cambodian rural population to become fully
participating members in the development process through decentralized governance.


       B. Immediate Objectives

The project is centered on the five immediate objectives:

1. To develop a participatory, demand-driven process of rural development, infrastructure
   planning and implementation.

2. To fund the construction of priority rural infrastructure projects identified through this
   participatory process.

3. To augment and institutionalize the capacity of the PRDCs.

4. To augment and institutionalize the capacity of the VDCs and CDCs.

To establish incentives for increased internal support for rural infrastructure projects through
financial and technical contributions of central government financial and in-kind contributions
and from local communities, while at the same time developing a sense of “ownership” of the
local projects.


       C. Outputs

The major outputs of the LDF project include the following:

1. The establishment of a pilot to test and further develop proposed LDF procedures in two
   provinces of Battambang and Banteay Meanchey (pilot phase), now expanded to include
   another three provinces.

2. The establishment and institutionalization of the LDF in a suitable agency at the provincial
   level following an evaluation of the pilot phase.
UNCDF Cambodia Final Evaluaton                                                            iv


3. The development of appropriate procedures and proposals for developing rural infrastructure,
   including: a program of training for PRDC, DRDC, CDC and VDC members that supports
   the institutional arrangements and procedures of the LDF outlined above; and
   institutionalization of the intergovernmental project planning and implementation process
   outlined above.

4. Physical outputs created by means of the LDF process.



3. PURPOSE OF THE EVALUATION

The overall objective of this final evaluation is to assist the Royal Government of Cambodia,
beneficiaries, UNCDF, and UNDP to:

♦ Assess the efficiency, effectiveness, relevance and impact of the project, and make
   recommendations for improvements where possible.

♦ Provide feedback to all parties to improve the policy, planning, project formulation, appraisal
   and implementation phases in the context of this next phase projections of the SEILA
   Program.

♦ Gather extensive lessons learned and operational recommendations that will feed into the
   future orientation and joint-formulation of the SEILA Program beyond 2000 with UNDP,
   SIDA and other possible partners. Recommendations should propose ways of ensuring a
   smooth transition of the project to the next phase and the further institutionalization of the
   project into the governmental structure taking into account the emergence of the “Commune
   Councils.”

Draw lessons that will improve the efficiency, effectiveness, relevance and impact of other
UNCDF-funded LDF projects.



4. FINDINGS AND ASSESSMENT OF RESULTS ACHIEVED


       A. Achievement of Immediate Objectives

As of August 1999, the LDF has supported the establishment of 1,144 locally elected Village
Development Committees (VDCs), 134 Commune Development Committees (CDCs), and five
Provincial Rural Development Committees (PRDCs), five Secretariats and five Ex-Com
Committees, all fully staffed by government civil servants.

Because of the lack of baseline data, it is difficult to measure impact; however, from what the
Evaluation Team (ET) has observed, the project has obtained the above objectives beyond
UNCDF Cambodia Final Evaluaton                                                              v


expectations. Projects are, on the whole, being completed in a satisfactory manner in less time
than planned; villagers are identifying and prioritizing their needs at the grassroots level; and
VDCs and CDCs are turning them into viable projects. Provincial government, through the
newly established PRDC, Executive Committee, and Secretariat, is establishing these local
development bodies and is assisting them in designing and implementing the projects.

The project has developed a financial management system that is providing transparency, good
tracking, and accountability. In addition, a Decentralized Development Fund has been
established that provides a system for other donors to invest development funds directly from
central government to the provinces. The monitoring and evaluation (M & E) of the projects is
carried out by technical staff of the government who have also established a bidding process for
contractors.

On the whole, local government is different than it was four years ago. Local government
institutions have enthusiastically accepted this bottom-up/top-down style of development.. They
are actively listening to people in the villages and communes and are engaging them in activities.

There are still many problems, particularly with lack of full village participation in the process.
Some people and groups are not engaged and sometimes are not properly informed.
Maintenance of completed projects is a big problem. SEILA/CARERE know about this and are
actively working to build up a viable system to address this. Capacities are still low compared to
other countries, and planning is still in the beginning stages at the CDC and VDC level.

To measure success as stated before, we compare the accomplishments of the present
government with the provincial government of the past. The difference in capacity,
participation, inclusion, communication, planning, and implementation are rather remarkable
when one realizes that the program has only been operating just a little over three years.


       B. Institutions and Capacity Building

The SEILA/CARERE/LDF project experiment has assisted the newly formed local government
institutions that have dramatically changed the way development is implemented, and in many
respects, the way local government itself operates. These structures provide the framework for
the Local Planning Process to function.

One of the most important institutional factors is the large increase in communication from the
villages through every level of government up to the PRDC. People all through the system have
been taught to brainstorm, to ask questions, and to state opinions. Higher government bodies
are responding through district integration workshops and through technical and facilitation staff,
whose roles are to work with the VDCs and CDCs. Communication has also been greatly
enhanced between sectors through these various structures. Government staff appears to have a
much clearer understanding of its roles and responsibilities and seems active and motivated with
a feeling of “ownership.”
UNCDF Cambodia Final Evaluaton                                                            vi


Extensive trainings at all levels have helped this process. In one year alone, 56,245 people (34%
women) were trained in 2,200 workshops in such things as planning, problem solving, finance,
water and sanitation, agriculture, management, gender, governance, health, infrastructure design,
and implementation among other things. Most of this training was for specific job-related
functions so they could put the skills learned into practice. Most important, however, was
learning by doing and working with counterparts in CARERE.

One of the biggest advantages of the new system is that funds stay under regularized financial
management procedures that are subject to internal and external controls. There are effectively
four levels of control, including the Ministry of Economics and Finance auditors, an internal
auditor appointed by Governors, external auditors who may be appointed by CARERE, and the
CARERE PSO financial unit.


       C. Local Planning Process

As of August 1999, SEILA/CARERE/LDF has provided support to 1,144 locally elected Village
Development Committees (VDCs), 134 Commune Development Committees (CDCs), and five
Provincial Rural Development Committees (PRDCs). Each year as the CDCs and VDCs gain
experience implementing the process, they demonstrate added maturity in needs assessment,
planning, and implementing projects, indicating that rural development structure is viable in
Cambodia and has gained acceptance as a viable model for decentralized rural development with
possibilities for replication nationwide.

Participation at VDC meetings is excellent at first when projects are being formulated but die
down after completion. Villages meetings are lively with much openness and sharing of ideas,
which is quite different from the old method of the village chief making decisions and then
telling people what to do. Women are usually the most vocal in the meetings and are often the
majority in attendance, but often are not listened to when final decisions are made.

Another change has been the simplification of PRA tools, which proved to be too complex as
introduced initially. These have been simplified into what are now referred to as data collection
tools. This change has reduced time and effort demands placed on villagers to complete tasks not
directly related to decision-making that can tax the interest of the villagers.

Integration workshops in one province have been reduced from two days to one, and
presentations by sector officials has been eliminated giving more time for CDCs to present their
project plans and receive feedback from the various sectors.

Allocation was originally based on the number of villages regardless of size. Now the move is
towards a combination of population and poverty, although the Project staff realizes that most
projects implemented rarely benefit the poorest of the poor. In fact, projects often benefit the
villagers who have the most land, best houses, draft animals and motorbikes.

The LPP is also working as a mechanism for planning and coordination of multiple sources of
investment in villages and communes, including LDF, sectoral funds, governmental department
UNCDF Cambodia Final Evaluaton                                                           vii


budgets, and outside support from NGOs. The LPP is open to other “financial windows” besides
SEILA, through the Decentralized Development Fund. This is supported by recent decisions by
the Royal Government, IFAD, and the World Bank to capitalize the LDF for new projects.

In light of limited LDF resources for individual villages and the expectation of the upcoming
commune elections scheduled for the end of the year 2,000, SEILA management is shifting the
focus of the program to the commune level. This raises questions about the sustainability of the
VDCs that have been created, as they need additional assistance to make them more
representative of the villages as a whole and more responsive to different groups within the
villages.


       D. Project Design, Implementation and Outputs

The institutions involved in the SEILA Rural Development Structure (Village Development
Committees, Commune Development Committees, and Provincial Rural Development
Committees), have overseen about 1,300 local projects valued at about US $3.8 million. Local
support in the form of cash contributions has totaled US $500,000, or 16% of total project costs.
Projects completed under the LDF represent approximately 8% of total CARERE activities. In
1999, the Royal Government made a significant contribution to the LDF of US $215,000. The
Government has now announced it will double this amount, to US $500,000, for the year 2,000.
Other official support has recently been announced by Ratanakiri province, which has decided
that line departments contribute 10% toward LDF projects, and Battambang province, which has
committed to contribute 20% toward a road maintenance fund.

The sectoral allocation of LDF funds has overwhelmingly favored transportation and irrigation
works, and to a lesser extent, school buildings and water and sanitation projects.
The technical quality of projects in general is good. Of those projects evaluated, over 80% are
acceptable, 18% have problems that can be fixed, and only 2% are unacceptable.

Competitive bidding was introduced in 1999 and is resulting in lower costs and higher quality for
projects. Average cost savings compared to estimate ranges from 6% to 10%. A pre-qualifying
system for all contractors has also been introduced

The original intent of the allocation process was to use a multi-criteria framework that included
project cost, number of beneficiaries, priority of problems or needs, cost effectiveness, time
frame, local contribution, and sustainability. This framework has proved overly ambitious in the
local context and has led to some confusion.

Allocation methods between provinces use various combinations of poverty and population
indicators. Some of these indicators are almost certainly colinear and therefore do not all need to
be included. Others are poor proxies for poverty, such as the number of children attending
school, which is also strongly affected by transportation problems. Using the percentages of
female-headed households and poor housing may be adequate, but further statistical analysis is
required to make a recommendation.
UNCDF Cambodia Final Evaluaton                                                            viii


The discretion available to PRDCs has led to wide variation in the amounts granted to
communes, in some cases double or half the average amount over a three- year funding cycle.
There is widespread agreement that simple criteria for allocation from communes to villages
work best and are most likely to be applied consistently. Within communes, a simple weighting
by population may be sufficient.

Maintenance of projects is a perennial problem for all rural infrastructure, and this is especially
true in Cambodia. The Evaluation Team visited many projects that were initially constructed to
an acceptable standard, but were deteriorating after only one or two years without adequate
maintenance. Roads become impassable after a couple of rainy seasons, fences around ponds
fall down, and water pumps break.

Data in the commune databases are not updated on a regular basis. This makes evaluation of
socio-economic impacts impossible.

The district integration workshops are working to improve the sectoral allocation of all sources
of funding, by making line departments consider work plans and sources of support for a
commune at one time.

The standard designs for projects have widespread support. More templates with greater variety
are needed to suit local material availability, soil conditions, and hydraulic conditions. There is
still not the capacity of technical staff to deal with unusual situations, and more training is
needed.

The most significant change from the previous financial system is that PRDC Executive
Committees will be responsible for management of all investment funds in the provinces,
whether they originate from the national budget or from external sources. They will also have
significant influence over line department activities through their financial oversight.



5. ASSESSMENT OF THE PROJECT DESIGN

The project was designed with the awareness that this was a policy experiment with many risks.
These risks were identified and weighed according to degree of risk. There had been some
progress in the pilot phase project in the two communes in each of the two provinces, but no
formal evaluation had been completed. There was, however, a strong indication that the
government staff had “bought in” to the process and that they were performing their duties with
enthusiasm and above expectations. In addition, the various people in the VDCs and CDCs were
performing well.

The initial design favored the typical NGO integrated rural development approach where much
time was spent in the villages completing PRAs and RRAs involving endless data gathering. It
was hoped that villages would not only understand much about data collection, needs
assessment, and project formulation, but also that more of a sense of community would be
established. This proved far too time consuming for the villagers and too expensive for the
UNCDF Cambodia Final Evaluaton                                                            ix


project. This approach has been modified, and a much simpler method is now being
implemented with more than adequate results.

The project was also started without establishing a formal monitoring and evaluation system.
The main phase document reflected this and pointed out the serious need for a comprehensive M
& E system to be developed and implemented. This has proved to be a serious problem, and it is
only in the past few months that this has been adequately addressed.

In other respects, the design has performed beyond expectations. The overall goal and
immediate objectives of the project have been achieved. The institutions created have been
strengthened, the LPP is thoroughly integrated throughout the system, and good quality
infrastructure has been implemented.

The project is costly, for it not only has to provide staff for capacity building, training,
monitoring, evaluation, and management advisory services, it also has to provide salary
supplements to government staff. Government staff receives (this is in addition to their regular
salaries which average about US $15 per month) US $80 per month (provincial level staff) and
US $40 per month (district level staff). In addition, motorcycles are provided, as staff are
constantly traveling to the various CDCs and VDCs. This raises great debates about the
sustainability of the project. It is perhaps the biggest issue of the project.

The Evaluation Team views Cambodia as unique from most other developing countries. It
therefore requires much more time and effort in capacity building, training, technical support,
and advisory services which are extremely time and people intensive and therefore, expensive.
This support is also necessary,especially when you are literally changing the way local
government operates and delivering vitally needed infrastructure at the same time.



6. POLICY IMPLICATIONS AND LESSONS LEARNED

The project demonstrates the following:
UNCDF Cambodia Final Evaluaton                                                              x


•   The top management of CARERE had years of experience working with Cambodians in the
    field, knew their language, their culture, the horrors of their experience, and appreciated the
    enormous constraints to development that existed in the country at the beginning of
    CARERE 2.

•   Local government can be changed to be participatory and decentralized when there is “buy
    in” at the top of the provincial government. The present governors of both Battambang and
    Banteay Meanchey, where the project started, enthusiastically supported the mission of the
    project, as did other influential people in the provinces including two former governors.

•   A successful local government project in one province (in this case two provinces) can lead
    to replication in other provinces and eventually receive support by national government.

•   Although operating with the same mission and strategic objectives, there were different
    methods of fund allocation and project selection that allowed for considerable adaptation to
    local situations.

•   Intensive training at all levels of provincial government was followed by opportunities for
    staff to implement what they learned, while at the same time receiving back-up technical
    support. This process not only dramatically increased the capacities of staff but also changed
    their behaviors, attitudes and work habits and in the end, it greatly assisted in changing local
    government.

•   With a clearly defined “mission” by all the top players of the project, and through training
    and close supervision, the mission was integrated among government staff. The project was
    therefore driven by a clear sense of shared goals and vision that was “bought” by the great
    majority of players.

•   Beneficiaries who saw the possibility of immediate benefits from government realized from
    the beginning that the participation asked for was real and not another of the many empty
    promises from government.

•   Government was involved from the beginning and was the actual implementor of the project;
    therefore, the project had ownership throughout government.



7. RECOMMENDATIONS OF THE MISSION


       A. General Recommendations

•   It should be made clear to CDCs and VDCs from the beginning what the government will do
    and where their responsibility ends as far as maintaining completed projects. CDCs and
UNCDF Cambodia Final Evaluaton                                                              xi


    VDCs should have carefully developed maintenance plans including budgets for maintenance
    of projects.

•   There appears to be a strong need to inform the NGO community of the real mission of the
    project and develop a public relations strategy that will more fully address this issue.

•   New strategies should be developed in the VDCs whereby more of the total village is
    represented and where more inclusion of traditional leaders is advanced.

•   It should be made clear from the start that after the first few VDC meetings that all meetings
    should be run by the VDC Chairperson and not the government facilitator. Also chairpersons
    should be trained at the CDC level on how to conduct participatory meetings and how to
    keep different groups involved.

•   Give villages a clearer explanation of the CDC and its roles and responsibilities as well as the
    LDF and how it works.

•   Villagers should be notified well in advance of all VDC meetings, and should always be
    involved in decisions on such things as where wells and ponds should be located, how
    contributions should be collected and how projects will be maintained.

•   The members of the SEILA team should consider spending the year 2000 consolidating
    current activities, bringing the main experimental aspects of the local planning program to a
    close and thinking about ways to work strategically, given existing constraints and demands.

•   LDF funding is due to end this year, while the SEILA/CARERE project is funded through
    the year 2000. The project needs LDF funding to take it through the year 2000.

•   Emphasis in the year 2000 should also focus on building training-of-trainers’ capacity within
    the facilitation teams at the provincial level.

•   SEILA/CARERE/LDF should decide on how many communes and villages they can
    practically bring under the LPP and LDF and complete strategic plans with time frames and
    budgets that address this.


       B. Future Project formulation

UNCDF/UNDP or other donors could provide a full time local government and decentralization
specialist for one year to work between SEILA, CARERE and national government to assist
them in the formulation of plans to take the project into 2,001 and beyond. The role would be
for a technical advisor in local government and decentralization and would require facilitation
skills. The person would live in the country, network between different agencies, serve as a
catalyst, and assist in the development of a three-to-five-year strategic plan to which donors
could respond. Some of the possible issues the person would deal with are:
UNCDF Cambodia Final Evaluaton                                                             xii


♦ The experience to date has been that the LPP process has been modified in each of the
     SEILA provinces to suit the local context. In the next phase of the project, it will be
     important to begin developing a general framework for national guidelines that still allows
     local flexibility.

♦ Rather than reduce present SEILA staff, it is possible that staff should be increased to
     provide key people to go into other provinces to train local leaders in the methodology.

♦ Reinforce the new Decentralized Development Fund at the national level that will be a
     depository from an expanded list of donors from which SEILA could work.

♦ With the establishment of the new Commune Councils, there is a need to identify and
     coordinate outside financial resources for development projects, including multi-sectoral, at
     the commune level.



8.    CONCLUSION

SEILA/CARERE/LDF has been an enormously ambitious project. It has also been costly. It has
attempted to almost remake a local government process and deliver vitally needed infrastructure
development at the same. Although there is scope for improvement and much hard work yet to
be done to institutionalize the structures and processes that have been created, the creation of this
system has been a major achievement and has for the most part achieved its ambition. The
SEILA/CARERE/LDF system has accomplished the following:

1. The project has more than achieved all of its goals and objectives and the outputs described
   in the main phase document.

2. It has assisted the governments in the five SEILA provinces in establishing a participatory
   decentralized local government planning and development system in a government/society
   known for its top down hierarchical style of rule.

3. It has implemented a precedent in local government in Cambodia whereby a certain
   percentage of government committee members must be women.

4. The system created has been formally accepted by the RGC and is looked upon by most
   government officials as the model for decentralized local government in all provinces.

5. It has had a major influence in the drafting of new local government laws and with the
   formulation of the Commune Councils in the year 2,000, it is seen as a move towards
   democratization of local government.

The major task of SEILA/CARERE /LDF is now to consolidate progress and to assist and work
with the Royal Government of Cambodia, as the government desires, and other donors in the
formulation of strategies and plans to move the SEILA/CARERE/LDF project forward into the
UNCDF Cambodia Final Evaluaton                                                           xiii


year 2,000 and beyond. In addition, to assist the government, if it so desires, in its move towards
decentralization in other provinces.


9. MEMBERS OF THE EVALUATION TEAM

The Evaluating Team is comprised of the following:

•   Robert Leonard, team leader, (a local government and institutional consultant who lived in
    Cambodia for almost four years), has trained all of the provincial Governors and First Deputy
    Governors and all of the top leaders in each of eleven provincial offices in management and
    planning.

•   Dr. Robert Guild, an infrastructure specialist and engineer who recently completed an
    evaluation of the Vietnam Rural Infrastructure Development Fund for UNCDF.

•   Cristina Mansfield, a community participation consultant with four year’s experience in
    Cambodia working with international and local non-governmental organizations (NGOs).
ANNEX I           TERMS OF REFERENCE

                            UNITED NATIONS CAPITAL DEVELOPMENT FUND

                                          TERMS OF REFERENCE

                                           FINAL EVALUATION



Country                               :               Cambodia

Project Number                        :               CMB/97/C01

Project Title                         :               Local Development Fund

Sector                                :               (0210) Development Strategies, Policies and Planning

Sub-sector                            :               Local Governance

Government Executing Agency           :               SEILA Task Force

United Nations Cooperating Agency     :               UNOPS

Total project cost                    :

Financing:
                  UNCDF               :               US$ 4,5million
                  UNDP                :               US$ 20,5 million (as of June 1999)
                  Others              :               US$ 35 million
                                                      (as trust fund and cost sharing contribution)

Actual UNCDF disbursement at evaluation:

                  UNCDF+UNDP+PPC      :               $$………..
                  Community           :               $ ………
                  TOTAL (June 1999)                   $……….

Approval Date                         :               April 1997
Starting Date                         :               August 1997
Completion Date                       :               31 December 1999
Evaluation Date                       :               July 1999




                                                  1
FINAL EVALUATION

This final evaluation has two purposes:

        i.      it is a standard evaluation of an ending project as per the rules and regulations of the UN
                Capital Development Fund; and

        ii.     it is an operational and strategic evaluation for the project to continue in a new phase, to be
                financed by various donors; i.e. it is to feed into the joint-formulation exercise for the SEILA
                Program beyond 2000.


PART ONE: GENERAL EVALUATION FRAMEWORK

I. Project Background

The overriding objective of the Royal Government of Cambodia (RGC), as expressed in its first Socio-Economic
Development Plan (SEDP) 1996-2000, is to achieve a fair, just and peaceful society and, through an
acceleration in the rate of economic growth, to raise the living standards of all Cambodians. Poverty alleviation
is the single most important goal.

To this end, the plan explicitly acknowledged the importance of investment in rural areas, recognizing that the
society and economy of Cambodia remain fundamentally agrarian. The plan also stresses the need to
restructure government with a view to achieving a professional public administration that is well-trained, efficient,
decentralized, politically neutral, and responsive to the needs and concerns of the people.

Underlying this new direction, the SEDP recognizes two important preconditions for development of the rural
areas – i. the need to rationalize the appropriate levels of responsibilities for local administrations (province,
district, commune and village) and establishing accountability and effectiveness in local level planning and
financing of development activities, and ii. the need to develop political and institutional support for
decentralization.

The Government’s experimental decentralization program, known as the SEILA program, is a key component of
the SEDP. The objectives of the SEILA program are to alleviate poverty and to promote peace by improving
local governance through intensive capacity building and technical assistance, broad based participatory
planning, financing and implementation of development projects.

The United Nations Development Program (UNDP) supports the SEILA program through the Cambodia Area
Rehabilitation and Regeneration (CARERE2) project. The current CARERE2 project follows the CARERE1
project (Cambodia Resettlement and Reintegration project) that was launched following the Paris Peace Accord
in 1992. The emphasis of the CARERE1 project was to support emergency, quick impact projects providing
basic rural infrastructure and essential services for resettlement. The CARERE2 project is designed to facilitate
the rapid and sustained shift from provision of rural infrastructure and essential services to intensive capacity
building of Cambodian institutions responsible for the sustained provision of these services and local
development.

The Cambodia Local Development Fund (LDF) project was formulated in August 1995 and a pilot phase started
in early 1996 in two communes each in Battambang and Banteay Menachey. The pilot evaluation was
undertaken in April 1997. This was followed by the formulation of a main phase in August 1996 and its
implementation period was commenced since early 1997 until 1999. The main phase covered five
SEILA/CARERE provinces namely Battambang, Banteay Meanchey, Siem Reap, Pursat and Rattanakiri.




                                                         2
The LDF projects have been carried out in more than 800 villages and 100 communes across the five provinces
under the area of 500 villages and 40 communes. The LDF supported 370 small-scale projects in the second
                                         1
cycle of project implementation in 1999.

The Cambodia LDF complements both the national and UNDP/CARERE policy objectives by investing in
decentralized delivery of basic social and economic infrastructure to create economic opportunities for the poor,
improve their access to services and enhance their human capital.

A. Pre-Project Situation:
Refer to Project Document

B. Project Rationale:

The interrelated rationales for the Cambodia LDF project in Battambang Province and Bantey Meanchey
Province are -

 i.  Infrastructure development is well known to play a critical role in poverty alleviation and economic
     development.
ii. During the past few years, the Government of Cambodia has placed an increasing emphasis on furthering
     rural development.
iii. The Government has embarked on a major program of decentralization and participation in the development
     decision-making process.
iv. Government resources to implement its intended programs of infrastructure provision, rural development,
     and decentralized planning and implementation of projects are extremely limited.
v. The establishment of the LDF is a critical component of UNCDF/UNDP efforts to reorient their broader
     program of activities in Cambodia.

In summary, the core rationale for the project is to assist the Government of Cambodia as it develops critically
needed rural infrastructure to support local economic development through a decentralized, participatory
decision-making and implementation process.


II. The Project

A. Development Objective:
To the alleviate poverty and spread peace in Cambodia by strengthening the bonds linking society to the
structures of the State and empowering the Cambodian rural population to become fully participating members
in the development process through decentralized governance.

B. The immediate objectives:
The LDF project has the following immediate objectives:

i.     To develop a participatory, demand-driven process of rural development, infrastructure planning
       and implementation to support projects that generate greater economic development and improve the
       quality of life for poor residents of rural areas;

ii.    To fund the construction of priority rural infrastructure projects identified through this participatory,
       demand-driven process;

iii. To augment and institutionalize the capacity of the PRDCs: to develop and facilitate local participatory
     information gathering and planning exercises; to coordinate and assist provincial technical department in
     supporting the development of locally initiated rural infrastructure projects; to evaluate infrastructure project
     proposals made by the VDCs/CDCs; to manage the disbursement of funds for the LDF infrastructure
     projects; to monitor and evaluate implementation of LDF infrastructure projects.


1
     UNDP/CARERE, Annual Project Report, 1 January 1998 to 31 December 1998, April 1999

                                                                   3
iv. To augment and institutionalize the capacity of the VDCs and CDCs (and eventually the DRDCs): to
    collect and analyze basic economic, demographic and social information at the local level; to identify and,
    given budget constraints, prioritize local infrastructure project needs; to work together with other VDCs and
    CDCs on joint projects when appropriate; to work with various types of project partners -- NGOs, provincial
    level technical departments, e.g., Department of Public Works, Department of Education, etc., and private
    sector firms -- to design, implement, operate and maintain local infrastructure and service facilities; to solicit
    funds from other sources that complement the resources of LDF, including the Social Fund2, UNICEF, the
    proposed UNCDF road program, and international NGOs that work in their areas; and

v. To establish incentives for increased internal support for rural infrastructure projects through:
   financial and technical contributions of the central government towards developing local infrastructure;
   financial and in-kind contributions to support rural infrastructure projects from local communities who benefit
   from these services, while also developing a sense of “ownership” of the local projects.

C. Outputs:
The major outputs of the LDF project include the following:

i.         The establishment of a pilot to test and further develop proposed LDF procedures in two provinces of
           Battambang and Banteay Meanchey (pilot phase), now expanded to include another three provinces – Siem
           Reap, Pursat and Rattanakir (main phase).

ii.        The establishment and institutionalization of the LDF in a suitable agency at the provincial level
           following an evaluation of the pilot phase;

iii. The development of appropriate procedures and proposals for developing rural infrastructure,
     including: a program of training for PRDC, DRDC, CDC and VDC members that supports the
     institutional arrangements and procedures of the LDF outlined above; and institutionalization of the
     intergovernmental project planning and implementation process outlined above. It is hoped that this
     process will be used for all local project planning in the SEILA framework, not just for those projects
     financed through the LDF.

iv. Physical outputs created by means of the LDF process, as an important tangible outcome of the
    decentralised development process in the Pilot Phase target areas.


III. Project Implementation Status

The documentation prepared by AIT and by the project staff provides a comprehensive overview of the project
implementation status. It contains among others:

i.              the local planning process, its revision and its effectiveness,

ii.             status of sub-projects implementation by SELILA/CARERE provinces,

iii.            capacity building activities of SEILA

(See in annex 1 list of documents produced by AIT)


IV. Purpose of the Evaluation

The overall objective of this final evaluation is to assist the RGC, beneficiaries, UNCDF and UNDP to

       2
           This is supported by the World Bank



                                                                 4
i.          assess the efficiency, effectiveness, relevance and impact of the project, and make recommendations
            for improvements where possible

ii.         provide feedback to all parties to improve the policy, planning, project formulation, appraisal and
            implementation phases in context of the new phase SEILA Program;

iii.        gather extensive lessons learnt and operational recommendations that are to feed into the future
            orientation and joint-formulation of the SEILA Program beyond 2000 with UNDP, SIDA and other
            possible partners. Specifically, recommendations should propose ways of ensuring a smooth transition
            of the project to the next phase, strengthening of the Participatory & Monitoring Evaluation (PM&E)
            system, and the further institutionalization of the project into the governmental structure taking into
            account the emergence of the “Commune Councils.”

iv.         draw lessons that will improve the efficiency, effectiveness, relevance and impact of other UNCDF-
            funded projects;

In evaluating performance, obstacles that are hindering the implementation or operations of the project should
be identified and solutions recommended.

V. Evaluation Methodology

The mission will be carried out in the following way:

i.          A briefing of the mission team or at least the team-leader at UNCDF headquarters. Relevant
            documentation will be provided to the team members prior to the mission.

ii.         As there has been several evaluation and study missions conducted recently that are related to this
            project, the evaluation team should utilise the data and information already available from these
            missions; including the UNCDF Donor Evaluation Cambodia Country Report (1999), the mid-term
            evaluation reports of the CARERE2 Project (1998) , and the independent monitoring and evaluation of
            CARERE reports support to the SEILA planning process (1998 and 19993)

iii.        Prior to the mission, the team leader should consult with the other team members and propose a
            detailed methodology on how to proceed with the evaluation; i.e. the feasibility of sample surveys,
            participatory learning and action/participatory rural appraisal (PLA/PRA), focus-group discussions, etc.
            This would include the tasks to be performed by the different team members. This proposal should
            include a time schedule for the mission and a table of content for the evaluation report. See PART II
            Detailed Terms of Reference.

iv.         Review of relevant project documents and files. This includes the reports listed in the Annex, particularly
            reports prepared by AIT. The AIT reports will be made available to the team leader electronically before
            the mission.

v.          Briefing with AIT will take place in Bangkok.

vi.         Briefing by UNDP/UNCDF, SEILA/CARERE and LDF project staff and review of the project files in the
            field.

vii.        Field trips and site visits to conduct the evaluation will be planned in consultation with UNDP,
            SEILA/CARERE and LDF project staff, to meet with the relevant project-related authorities and the
            beneficiaries/users, as well as population groups outside the project areas. These should include
            members of VDCs, CDCs, DDCs and PRDCs, women, leaders, and the poor, SEILA/CARERE
            management, national and technical staff, NGOs and donors. Wherever possible, evaluation data
            should be disaggregated by gender.

3
    CARERE Evaluation Report on the Third Cycle of LPP (1999) is expected to be ready by the end of this month or in coming August.

                                                                        5
viii.           Debriefing session with Government UNDP/UNCDF in Phnom Penh, and other SEIILA donors.

ix.             Debriefing to UNCDF New York by the team leader.



VI. Organization of the Mission

A. Composition of the mission:

The following profiles of the Consultants are suggested:

i.              Team Leader: Specialist in rural planning and decentralization with several years of working experience;
                preferably with working experience in Asia;

ii.             Community participation expert, with extensive experience in Cambodia.

iii.            Infrastructure expert


B. Duration of the mission:

The mission will start on _______ 1999, for the duration of approximately seven work-weeks, including time
required for briefing and debriefing in New York and Bangkok.

This will include:

        i.          briefing and orientation of 1-3 days at UNCDF headquarters for the mission leader;
        ii.         1 day briefing at AIT;
        iii.        3 weeks of evaluation work in the field;
        iv.         1 day for Evaluation Mission Wrap-Up meeting;
        v.          1 day de-briefing at AIT;
        vi.         2 weeks report drafting;
        vii.        1 day debriefing by the mission leader at UNCDF headquarters;
        viii.       3 days to incorporate feedback received during debriefing (NY).


VII. Reporting:

The mission should discuss its main findings and recommendations with the UNDP Resident Representative,
government authorities, and other project partners concerned. Ideally, time should also be allowed for
discussion of findings in the field with the project beneficiaries.

At the end of the mission, an Aide Memoir should be drafted to be the basis of discussion at the Evaluation
Mission Wrap-Up meeting. UNCDF should receive a copy of the Aide Memoir as well prior to the meeting. While
the consultants are free to raise any subject relevant to the evaluation of the project, the mission is not
empowered to make any commitments on behalf of UNCDF.

The comments of the Government, the UNDP Resident Representative and other relevant donors on the Aide
Memoir and at the Wrap-Up meeting should be incorporated or addressed appropriately in the draft evaluation
report.

Within two weeks after completion of the field evaluation in mid-September, five copies of a draft report,
including the "summary of project evaluation", should be submitted to UNCDF headquarters for review and
comments prior to the debriefing of the evaluation mission and finalization of the evaluation report. This should


                                                               6
take place sometime in mid-October. The Government, the UNDP Resident Representative and other donors
are expected to forward their comments on Evaluation report to UNCDF.

Upon receipt of all these comments, the team leader will finalize the report and present 10 bound copies and 1
electronic copy (in WordPerfect 5.1 or Microsoft Word 6 format) of the final evaluation report to UNCDF. UNCDF
will distribute the final report to all parties concerned.

VIII. Mission costs and financing:

The costs of the mission will be charged to CMB/97/C01 budget line 16.02 (mission costs). The final payment of
fees will be paid only upon acceptance by UNCDF of the final evaluation report with 50% of the agreed amount
upon acceptance of the draft evaluation report.


IX. Approach to the Evaluation Report

The evaluation report should follow the UNCDF. Section A, chapters I and II, should be a record of the facts,
without reflecting the author's opinions or assessment; the latter should be included in section B. In the
evaluation report, the consultants are free to raise any subject relevant to the evaluation of the project.


PART TWO: DETAILED TERMS OF REFERENCE

I. Contents of the Evaluation Report
The evaluation report should include the following items:
i.      Table of contents
ii.     An Executive Summary, to include an overview of the report, and a summary of the main findings and
        recommendations.
iii.    List of abbreviations
iv.     A project data sheet, providing key facts and figures on a single page
v.      An Introduction, briefly stating the purpose of the mission, composition of the evaluation team, a
        schedule of activities carried out, the methodology used, and the structure of the report.
vi.     Chapters that cover the topics and issues outlined below
vii.    List of persons interviewed
viii.   List of documents used in the evaluation
ix.     An Evaluation Summary, as a 4-5 page annex to the main report. This is distinct from the Executive
        Summary, and should serve as a self-contained summary that may be read without reference to the
        main report. The evaluation summary should cover:
            a. Basic project data
            b. Background of the project
            c.   Description of the project
            d. Purpose of the evaluation
            e. Findings of the evaluation mission
            f.   Assessment of the project design
            g. Policy implications and lessons learned
            h. Recommendations of the mission


                                                       7
               i.   Members of the evaluation team


UNCDF will provide an example of an Evaluation Summary to the team leader.


II. Factual Presentation of the Information Necessary for Project Evaluation

A. Evaluation tasks pertaining to the original project design and situation
The evaluation report should include a brief description and overview of LDF from inception to the present. It
should indicate:
i.         The country and sector, the situation existing at the time the project started, selected data to
           illustrate prevailing conditions in the areas targeted by the project, the origin and evolution of the
           project;
ii.        The project rationale, the substantive approach, the development objectives, immediate
           objectives; expected results (outputs), activities, project inputs, implementation arrangements,
           costs and financing, including the Government's funding commitments; plan of operations; and
           arrangement for monitoring and evaluation (M&E); The mission is requested to describe the actual
           change in the mode of implementation compared with the project document. The relevant information is
           contained in the Technical Review Mission Reports, prepared by AIT.
iii.       Risks identified at the time the LDF was designed.

B. Evaluation tasks pertaining to project implementation results
Describe all facts that reflect the status of the project. These may be organized into 3 categories –
results (outputs), processes, and impact.

i.     Results (or outputs) should be presented with regard to delivery of all inputs (i.e. physical and technical
       assistance), timing, quantity and quality, activities, outputs (both physical and process oriented), costs,
       utilization of outputs, maintenance, project operation, beneficiaries, etc.

ii.    Project processes should be documented, describing the procedures, activities, and timing, of the different
       project stages; covering formulation, inception, implementation, monitoring and evaluation, operations and
       maintenance, including the operational processes therein; i.e. hiring of staff, training, facilitation of meetings,
       selection of projects, procurement, contracting, payments, etc.

iii. Lastly, changes in quality if life indicators should also be presented as basis for the socio-economic
     impact analysis to be done in the impact evaluation section.

It is important that, wherever possible, all data gathered should be disaggregated by gender, socio-
economic and social groupings.



III. Assessment of Results and Status of Implementation
The evaluation should be concerned with the broad questions of whether the LDF project:
i.         is indeed improving the way in which Commune–level capital development programs are financed,
           planned and implemented,
ii.        has resulted in more relevant and sustainable rural infrastructure projects and
iii.       provides an institutionally viable and financially sustainable approach to decentralized rural
           development.




                                                             8
A. Evaluation of Results (Outputs)
In this section, evaluators should describe outputs relative to targets, and assess the extent of progress in
achieving the immediate objectives as stated in the project document.
Immediate Objectives

i.    To develop a participatory, demand-driven process of rural development, infrastructure planning
      and implementation to support projects which generate greater economic development and improve the
      quality of life for poor residents of rural areas.

The first immediate objective of the LDF is essentially a capacity building activity for participatory planning, for
which the evaluation team should:

1.     describe/highlight the participatory components in the local planning process,
2.     describe the methods adopted by the LDF to promote community participation at all stages; i.e. planning,
       implementation, monitoring and evaluation, operation and maintenance. This should include specifically
       documentation of the establishment of “user groups”, village development committees (VDCs), commune
       development committees (CDCs), and their respective roles. The timing and duration of each activity
       should also be documented in order to assess timeliness and speed.
3.     enumerate the technical assistance activities undertaken through the project to support the process,
       including training of facilitators, PRAs, follow-up meetings, including monitoring activities.
4.     gather quantitative information on – the estimated number of community members involved in each
       training, and in each stage of the project cycle - planning, implementation, monitoring and evaluation, and
       operations and maintenance activities,
5.     gather and assess qualitative information on the extent of community participation at each stage of
       planning, implementation, m&e and operations and maintenance at the village level, with attention to
       actual decision-making.
6.     evaluate the costs and benefits of community participation as practiced in the LDF, taking into account the
       costs involved (such as the time and effort of community members, village officials, the staff of line
       agencies and the LDF project team), and the benefits generated (particularly the impact on project
       selection, project design, community contributions, and timely implementation of sub-projects)
7.     assess the adequacy of institutions and techniques for community participation adopted or introduced
       under the LDF program;
8.     assess the effect of participation in LPP on the solidarity within and between communities; i.e. if
       participation has improved the solidarity of the community or promotes “local -local dialogue”.
9.     assess the attitudes of the project staff and relevant local government staff and village authorities towards
       community participation, women participation, and participation by different socio-economic groups.
10.    assess the extent to which women took part in these activities, and identify constraints to their
       participation; with attention to participation in actual decision-making
11.    assess the extent to which the poorest population took part in these activities, and identify constraints to
       their participation, with attention to participation in actual decision-making.
12.    indicate variations in the level of community participation among different vi llages or communes, and
       identify factors which affect participation
13.    sample and assess the difference across communes or provinces in terms of technical skills and
       participatory attitude of their members;

ii.      To fund the construction of priority rural infrastructure (as well as non-infrastructure) projects
         identified through this participatory, demand-driven process;



                                                          9
The second immediate objective of LDF may be measured by the improvement in the quantity and quality of
local basic infrastructure and services in the project area. This should include the “non-infrastructure projects”
that have been funded by the LDF. Since no baseline data was collected at the start, evaluators may have to
survey beneficiaries’ memory of pre-project situation and their perception of change and improvement since.
The evaluators should assess -
1.     the quantity of LDF funded projects and joint projects – planned and actual, completed or underway,
       including qualitative specifications which will allow for qualitative assessment; e.g. x km of x type/category
       of roads, x number of x type of wells providing x amount of x quality water, x number of clinics providing x
       type of health services, etc.
2.     the number of projects completed or underway, relative to targets, and explain difference
3.     the effect of projects: what are perceived benefits and actual effects of different types of projects
4.     the relevance of the “objects of financing” (both infrastructure and non-infrastructure) covered by LDF and
       the potential and constraints for a widening of the LDF scope to local economic development activities. the
       technical quality of projects using a quality scoring/ranking system and comment on overall performance
5.     the impact on quality of regulated construction standards, local procurement, community contribution and
       community monitoring of construction activities.
6.     the cost effectiveness of the projects produced under the LDF program. The impact on cost of regulated
       construction standards, local procurement, community contribution and community monitoring of
       construction activities.
7.     the cost effectiveness of the projects compared with other service providers, i.e. sectoral departments,
       NGOs.
8.     the economic efficiency of the projects, and comment on overall performance, including identification of
       common factors determining efficiency/inefficiency

9.     the timeliness in implementation and completion of the projects and comment on overall performance and
       identify common factors
10.    the estimated number of beneficiaries for each project
       11.   the general satisfaction of users; i.e.
        a. % of people having used the projects or services, by type and by gender
        b. % of people having used the projects or services, by type and by gender, and satisfied with the
           quality
        c.   % of people having used the projects or services, by type and by gender, who are satisfied with the
             timeliness of the delivery
        d. % of people having used the projects or services, by type and by gender, who are satisfied with the
           cost of the service

iii.     To augment and institutionalize the capacity of the PRDCs (and DDCs): to develop and facilitate
         local participatory information gathering and planning exercises; to coordinate and assist provincial
         technical department in supporting the development of locally initiated rural infrastructure projects; to
         evaluate infrastructure project proposals made by the VDCs/CDCs; to manage the disbursement of
         funds for the LDF infrastructure projects; to monitor and evaluate implementation of LDF infrastructure
         projects.

In assessing progress toward this objective, evaluators should:

1.     report on the establishment of PRDCs and DDCs, their composition, role, responsibilities and activities,
       and describe the nature of technical assistance provided to them



                                                          10
2.          assess the capacity of PRDCs and DDCs to fulfill the tasks assigned to them, particularly with respect to
            planning, technical assistance, financial management, auditing, coordination, facilitation, monitoring and
            evaluation and other collaboration among the CDCs and VDCs, promoting and executing joint projects,
            elaborating local development plans, and generally mobilizing resources for local development
3.          gather information through participatory methods on the general satisfaction of committee members with
            the type and quality of activities (especially training) received through the LDF
4.          gather information from the supervisors or superiors of staff and committee representatives on their
            perception of improvement or changes in performance of staff members or representatives since training.
5.          gather, through participatory methods (focus group discussions, PRA, etc.), the general satisfaction of
            community members with the performance of PRDC and DDC members and local government officials,
            with focus on perceived change in quality since project began.
6.          in addition, the evaluation team should attempt to identify factors that affect the performance of the PRDCs
            and DDCs in carrying out their roles.

iv.            To augment and institutionalize the capacity of the VDCs and CDCs (and eventually the DRDCs):
               to collect and analyze basic economic, demographic and social information at the local level; to identify
               and, given budget constraints, prioritize local infrastructure project needs; to work together with other
               VDCs and CDCs on joint projects when appropriate; to work with various types of project partners --
               NGOs, provincial level technical departments, e.g., Department of Public Works, Department of
               Education, etc., and private sector firms -- to design, implement, operate and maintain local
               infrastructure and service facilities; to solicit funds from other sources that compleme nt the resources of
               LDF, including the Social Fund4, UNICEF, the proposed UNCDF road program, and international NGOs
               that work in their areas;

In assessing progress toward this objective, evaluators should:

1. report on the establishment of VDCs and CDCs, their composition, role, responsibilities and activities, and
   describe the nature of technical assistance provided to them

2. gather quantitative information on the number of committee members participating in LDF activities by type
   and duration.

3. assess the actual performance of LDF in creating capacity for planning and project management at the
   commune and village level.

4. document the types and quantity of technical assistance provided to VDCs and CDCs by support agencies
   (NGOs, technical departments, etc.) and assess the quality from perspective of VDCs and CDCs.

5. report on the amount and source of funds raised from other sources that complement the LDF, describing
   the process by which the different funds were accessed

6. assess the capacity of VDCs and CDCs to perform their function, collect service charges, cover their costs
   and maintain equipment, and indicate levels of consumer satisfaction with new or improved services

v.             To establish incentives for increased internal support for rural infrastructure projects through:
               financial and technical contributions of the central government towards developing local infrastructure;
               financial and in-kind contributions to support rural infrastructure projects from local communities who
               benefit from these services, while also developing a sense of “ownership” of the local projects.

1. document and compare the terms of co-financing laid down by the project with the actual amounts and
   proportions of government and community contributions to the cost of projects and services delivered, and

      4
          This is supported by the World Bank



                                                               11
    identify factors, which account for variations in the level of community support, and evaluate the policy
    implications of the results.

B. Evaluation of Implementation
This section of the report should be addressed to an evaluation of the means, processes and procedures used
to implement the LDF, and its overall performance in terms of economic efficiency, equity, transparency, and
effective management in delivering outputs. The evaluation should assess factors, both internal and external to
the project, which have contributed to or limited the efficiency of the project.
The evaluation should cover the performance of the project staff in operating the Local Development Fund and
in carrying out its capacity building activities. This includes specifically, review of LDF process, the local
planning process, the implementation cycle, the operation and maintenance, monitoring and evaluation, the
project management and administration, as well as special considerations and sustainability and
institutionalization.

1. The LDF Process
Based on the CARERE hypothesis and related to the use Local Development Fund, the evaluation should
assess the following qualities of the LDF:

i. The Local Planning Process and Implementation cycle
The Local Planning Process (LPP) constitutes a large part of the implementation cycle for sub-projects. Briefly,
the objectives are to set up an institutional framework for village cooperation, to generate a portfolio of sub-
projects, and to strengthen community participation in local decision making.
The five main steps involve setting up the institutional framework (VDCs, CDCs, PRDCs and support staff, as
well as linkages to local technical support agencies), selecting projects (including PRAs), detailed preparation of
projects (design, costing, permits), contracting (tendering, bidding and selection), and project execution.
Based on the revised LPP introduced in early 1995, the recent assessment undertaken by AIT, data available
from the Management Information System (MIS) developed by CARERE, and through a more in depth analysis
of a limited sample of ongoing or completed sub-projects, the evaluators should review and assess the
efficiency and effectiveness of the decentralized provision of rural infrastructure promoted by the program.

This should include an assessment of:

1. evaluate the validity of the objectives of the LPP, and the purpose, need and usefulness of each step in
   the process in addressing these objectives
2. assess the outcomes of the LPP in terms of the choice of projects, particularly joint projects, and their
   relevance in achieving the objective of the LDF; i.e. the relevance of the investments selected through the
   LPP for poverty alleviation.
3. using data from a sample of projects, estimate the length of time needed to complete each step and the
   implementation cycle as a whole
4. identify common bottlenecks and sources of delay in the process, and assess the effectiveness of actions
   taken (or that might be taken) to eliminate bottlenecks and reduce delays
5. evaluate the quality of planning generated by the process (such as village and commune development
   plans, the poverty assessment data base, diagnostic studies, PRA reports, project proposals and designs),
   identify areas where these might be improved, and indicate the kind of technical assistance required and
   who might best provide this
6. the quality and appropriateness of the SEILA-required documentation and procedures.
7. assess the quality and relevance of Commune-level diagnostic and strategic planning information provided
   by provincial/district planners and facilitators to Commune-level decision makers, for problem analysis and
   sub-projects selection;


                                                        12
8. evaluate the way in which the projects are appraised and approved in terms of equity.
9. assess the gender equity considerations in the LPP in light of gender disaggregated information on
   participation rates, gender specific barriers to participation, and qualitative analysis of women’s participation
   in actual decision-making in all stages.
10. assess the appropriateness, flexibility and efficiency of the current procurement arrangements for the
    implementation of the small scale rural infrastructures supported under the LDF program, including the
    procurement issues raised by community contribution of labor and materials.
11. assess the use of contracting approaches to projects procurement. The assignment of owners’,
    supervisors’ and suppliers’/contractors’ responsibilities under the program. Who does what? What degree of
    transparency?
12. assess the impact of decision-making techniques/procedures introduced by LDF on the actual outcome
    of the local planning processes and on the institutional strengthening and capacity building for decentralized
    rural development;
13. describe the relation between the planning process introduced under LDF and the use of resources other
    than those of the LDF account. The current and potential extent, scope and modalities for negotiation of
    additional support to Commune-level investment programs, from other “windows of financing” (provincial
    / national programs, NGOs, other donors’ programs)
14. local contribution policy review and feedback; e.g. review the impact of the cash contribution requirement
    (3%)from the Commune, especially in terms of participation of the poorest, and explore alternative resource
    mobilisation approaches
15. describe the relationship/link between the LPP and the Communes’ budget preparation and execution
    process.
16. assess the capacity of local owners, technical supervisors and contractors to perform the roles called for in
    the LDF arrangements and the performance of CARERE in strengthening such capacity.
17. consider the context of the planned commune council law, suggest what is or what should a commune
    plan look like in the future.
18. recommend exit strategies for old communes in 2000 (communes that have completed their 3 year cycle
    and we can not continue to finance given the level of funds available in 2000).
19. identify Key Lessons learnt as appropriate; such as:
    §    Modification of the planning techniques (i.e. PRA, indicators of poverty assessment data base system
         and decision-making criteria in project selection) and outputs (village and commune development
         plans)
    §    Proposal for IPF-distribution across communes
    §    Modifications in implementation of projects;
    §    Identification of Outstanding Constraints and Potentials
    §    Plans for future Capacity Building

ii. Decentralized Financing Procedures
The evaluators should review the arrangements for funds disbursement (direct provincial disbursements or
replenishment of Commune accounts), accounting and auditing, in terms of efficiency, accountability and
transparency, and the local capacity issues associated with them.
The evaluators should assess:
1. The degree of institutionalization of the LDF financing system;
2. the management modalities of the LDF account
3. the clarity of the rules of access to the LDF resources and the extent of dissemination of information on
   such rules to the prospective beneficiaries.
4. the accountability, efficiency and degree of transparency in the allocation of the financial resources


                                                        13
5. the accountability, efficiency and degree of transparency in the disbursement-payment cycle;
6. the long-term potential for central and local government capacity for accounting and auditing;
7. the cost effectiveness of the LDF process, in terms of time as well as financial expenditure relative to the
   amount of funds delivered.
8. considering the strong and weak points of the current LDF financial management, make recommendations
   for a national decentralized financing facility (and its management and disbursement procedures) to
   provide donors with a realistic instrument to co-finance RGOC block grants to Commune Councils.

iii. Allocation

Evaluators should assess –

1.    the way in which the resources are allocated (the Communes targeting formula) in terms of rationality
      and equity of the allocation formula.

2.    the relevance of the “LDF and Commune database” for the construction of a formula, for geographic
      poverty targeting.

3.    the sources of funds contributed so far through the LDF, the use of funds by sectors, the allocation by
      province, district, commune and village, and the level of per capita funding for each, and assess the
      implications

4.    the sectoral allocation of funds for social services delivery, environmental protection and economic
      development, etc. Patterns of investment: relative preferences for social or economic infrastructure, for
      village-level or Commune-level investments.

5.    the shift in focus of level for location of projects; i.e. the main focus in the beginning of the LPP was at the
      village level, with the expectation that this would gradually shift to commune level. To what extent has this
      happened/is this happening and what would be effective measures to stimulate this.

2. Capacity building
Here, evaluators should address process and efficiency issues related to the delivery of capacity building
outputs to the clients – the community, the local authorities and other relevant clients, including a review of the
topics to be covered and the means or instruments of capacity building.

The evaluators should review the programme’s approach to capacity building, in terms of targets, areas and
resources as specified in the program’s design. The capacity building component may be broken down into
three categories – capacity building of community to participate in LDF (related to immediate objective i),
capacity building of the local authority to manage and implement the LDF, and capacity building of other
relevant actors (related to immediate objectives iii and iv).

The evaluators should –

1.    identify the main goals of capacity building under LDF and evaluate the validity of these goals

2.    assess the content, materials and organization of training activities conducted under the project;

3.    identify the various client groups targeted for capacity building (including provincial and district facilitation
      teams, technical support staff as well as administrative/financial staff and members of VDCs and CDCs)
      and estimate the relative level of effort devoted by the project team to each of these clients

4.    assess the benefits derived from strengthening the capacity of each of these clients relative to the level of
      effort involved, identify weaknesses among the client groups, and determine how efforts might be directed
      more effectively


                                                          14
5.    assess the overall cost of training and other capacity building activities and compare with available
      comparators

6.    assess the performance of the training institutes (AIT, CARERE staff, local NGOs) and the scope for
      continuation/expansion of their services, versus the training functions that can be internalized by
      CARERE.

7.    assess the ability and effectiveness of different kinds of agents for capacity building, and options for
      making better use of available resources for this purpose. If available and appropriate, identify new
      agencies that might be used, for capacity building. This should include an assessment of the potential and
      actual role that the emerging private sector could play to address the capacity problems of the local
      (District and Commune-level) public sector.

8.    identify the main topics covered for each client group (such as PRAs, the local planning process,
      contracting and procurement, accounting and financial management), evaluate the utility of these topics
      (and possibly other subject areas not covered) in light of the goals for capacity building, and determine
      how efforts might be better directed

9.    indicate the different means and instruments used for capacity building in each subject area (such as
      manuals and guidelines, workshops, and on the job training), assess the relative efficiency and
      effectiveness of these methods, and indicate what improvements might be made

10.   assess the extent to which LDF capacity building efforts have been shaped by the needs of the project
      (supply driven) versus the needs of clients (demand driven), and indicate why and how the balance
      should be shifted one way or the other, and how this might be accomplished.

3. Operation and maintenance
Since one of the objectives of LDF is to improve the quantity and quality of infra structure and services in
underprivileged rural areas, a key issue in the longer term concerns the operation and maintenance of these
facilities once they have been built or established.
In this regard, the evaluation team should:
1.    evaluate typical arrangements made under LDF for operation and maintenance of different types of
      infrastructure, and the effectiveness of these arrangements in practice; evaluators should inspect a
      sample of projects for infrastructure and services financed through the LDF in earlier years to determine
      the current condition of physical infrastructure and the quality of operations and maintenance services
      being provided (particularly staffing, equipment and supplies
2.    assess the sustainability of infrastructure projects with current arrangements for operation and
      maintenance, detailing the sharing of responsibility between Commune Authorities, provincial/district
      agencies, and communities.
3.    compare the maintenance of LDF funded projects with those carried out by other service providers.

4. Monitoring and Evaluation
The project document for LDF includes detailed plans for a system of monitoring and evaluation, but in practice
these plans have only partially been implemented.
In addressing this subject, the evaluation team should:
1.    summarize the monitoring and evaluation plans and assess them in terms of their purpose and utility, and
      the feasibility of implementing them
2.    indicate what steps have been taken for the implementation of these plans, and explain if not or only
      partially implemented; describe the procedures adopted in practice to monitor and evaluate project
      activities, particularly the design of a management information system, and assess these procedures and
      the quality of outputs generated
3.    assess the shortcomings of the current monitoring and evaluation applied by the project;

                                                          15
4.   assess the degree of central and local monitoring of the implementation process; more specifically, the
     degree of provincial/district monitoring of the implementation process (procedural and quality controls).
     What is needed? What is appropriate?
5.   clarify the objectives of a monitoring and evaluation system for the project in light of the current situation,
     outline the components of this system, and determine what immediate actions should be taken in
     readiness for the next phase of LDF, to improve monitoring and prepare the ground for future evaluation
     studies.
6.   review and assess the “Management Information System (MIS)” developed by LDF to track the LPP
     and projects cycle.
7.   outline the action/research program which would be necessary to address the above questions and the
     type of MIS needed to support such a program; outline a Logical Framework for the next phase and
     suggest a limited set of Key Performance Indicators (KPIs) relevant for assessing and evaluating the
     (future) performance/impact of the project; including the set of indicators to monitor the performance of the
     emerging Commune Councils and which will facilitate management by Provincial authorities.
8.   suggest the key steps to be taken to establish a Participatory Monitoring & Evaluation (PM&E) system
     under which beneficiaries will monitor and evaluate the project’s impact themselves.

5. Project management and financial administration
As mentioned earlier, the next phase of the LDF will be jointly formulated by UNDP, SIDA and other donors. In
light of their concerns, mission members should pay particular attention to evaluating the management and
administration of the project. Among other things, they should:
1.   describe the current institutional structure for managing and administering the project, indicating the
     specific tasks and responsibilities of each institution at the national, district and local levels, including
     UNDP, the LDF team, PRDCs, CDCs and VDCs.
2.   describe the actors involved and the mechanisms for key decision making concerning project policies,
     the allocation of funds, and the selection of micro-regions and sub-projects, and evaluate these
     mechanisms in terms of transparency and accountability
3.   describe and assess the process and procedures for tendering, bidding and selecting contractors for
     construction works
4.   describe the procedures for the subsequent steps of supervising construction, making good
     deficiencies, technical clearance and certifying completion of works, and evaluate these procedures, the
     competence of supervising engineers, and the quality of site supervision
5.   describe and evaluate the quality of documents required or produced during these procedures (such as
     tender documents, bids, contracts, and certificates of completion)
6.   describe and evaluate the system for disbursement of funds and payment of contractor bills, including
     documents to support requests for payments, monitoring bills of quantities supplied for construction, and
     the nature and quality of financial controls.
In all of these cases, evaluators should identify problems encountered, and assess measures taken to resolve
these problems.

C. Evaluation of Impact
In order to assess impact, the evaluators should gather socio-economic indicators from all micro-regions served
by LDF as well as other areas for comparison. Specifically, assessment should be made of changes since
project implementation, in –
1.   poverty indicators: such as, expenditure, assets, housing type, etc,
2.   health indicators; such as morbidity rates, mortality rates (infant and maternal), malnutrition incidence,
     changes in diet, etc,



                                                        16
3.   access to knowledge indicators, such as literacy, primary school enrolment, survival rate up to grade 5,
     access to information, access to power brokers, government officials, etc.
4.   environmental indicators, such as soil fertility, water resources, land cover, etc.
Comparison between areas should allow for the determination of the extent to which the project has in fact been
targeting underprivileged rural areas. To do this, evaluators should –
1.   Identify the beneficiaries (planned versus actual);
2.   Determine whether the project has reached the expected target beneficiaries - if not, why not?
3.   Assess possible current or future constraints to reaching the intended impact.

D. Project preparation and design

Analyze the appropriateness of the project preparation (baseline data, feasibility studies), and assess the design
and the quality of the formulation of the project, based on the Project Agreement. Were all the necessary
components/elements taken into account? Specific attention should be given to the analysis of and accounting
for the institutional and capacity issues in the project design.

E. Overall Project relevance and effectiveness

The mission should:

1. Compare the achievements of the project to what was expected at the time of its formulation, and
   determine whether the project has been effective; Assess how well and to what extent the project objectives
   are likely to be achieved. Are they realistic? Are they still valid? Is the project designed in the best possible
   way to achieve the objectives set forth?

2. Assess formulation assumptions and verify if these are still valid

3. Assess the extent to which the outputs were well targeted and/or designed to achieve the project's
   immediate objectives.

4. Is the project cost-effective?

5. Are the results obtained in line with the national development objectives?

6. What are the possible alternatives/ improvements that could have been or could be applied?

7. Present a clear presentation of the hypothesis and assess the extent to which it has been verified.

8. Assess the project's suitability and effectiveness as a "guide" for similar activities within and/or outside the
   country, particularly with regard to its influence on national policy and/or practice.

F. Critical and Strategic issues

The mission should assess and report any features that are critical to ensure the sustainability and replicability
of the project activities.

1. Sustainability
The evaluators should review the current situation in terms of potential sustainability of the LDF and identify the
conditions for financial and institutional sustainability of the fund, beyond the current program’s duration.
Specifically, the evaluators should assess -

1. the level of capitalization and replenishment of LDF so far, by both UNCDF and the Provincial Authorities
   and the long-term sustainability of the current level of LDF capitalization.


                                                        17
2. the arrangements being made by the Provincial Authorities for a sustainable replenishment of the
   account, beyond the period of UNCDF financial assistance.

3. the arrangements being made by the provincial authorities for the long-term management of the LDF
   account (sustainability of Rural Development Structures or other institutional arrangements)

4. the forms of co-financing and the level of additional local resources mobilization (from other provincial
   programs, Communes’ own resources and community contributions) induced by the availability of LDF
   resources.

5. whether or not other donors or NGOs in the five CARERE provinces have adopted and/or modified
   LPP/LDF processes in their work.

6. any other institutionally and financially sustainable structures the State of Cambodia should set up to
   facilitate the Commune-level adoption of the participatory planning and improved management procedures
   introduced by the LDF program (as appropriately revised).

2. Strategic evaluation:

The overall strategic analysis should refer and take into account the main points and recommendations made in
the UNCDF review of LDF (August, 1997) and the Seila Task Force recommendations on Local Development
Funds.

The strategic evaluation of the project will assess the opportunities and constraints for the integration of the
LDF project, in terms of its institutional and technical features with the larger national policy and institutional
environment set by the Public Administration Reform (PAR) process.

Specifically, the evaluators should appraise the ongoing initiative to set up a Government Task Force for the
mainstreaming of the decentralized financing and planning procedures developed by the LDF program, within
the system of local fiscal transfers and Commune-level capital programming and budgeting.

How should the LPP introduced by the program, evolve (in terms of both its institutions and techniques) to
qualify as a procedure that may be nationally mandated for the Commune Councils to follow for (1) strategic
planning, (2) multi-year programming and (3) annual budgeting of local development expenditures.

As part of this assessment, evaluators should also gather information on what are the main sets of views about
the change processes relating to the project held by main partners and agencies (donors, implementing
consultants, local project office, community groups, local line departments)? How do these views differ from and
what do they say about the necessary and sufficient conditions for the project advancing? What are the paces at
which various change processes are occurring (at the various levels including population, local official)? What
do local communities think of the project, its methods and rationale?


G. Findings, recommendations and policy lessons learned

The mission should be requested to:

1.   State the findings of the evaluation, taking into consideration both the successes and failures of the
     project. Special issues that need to be addressed as a result of project activities, or which will improve the
     project's effectiveness, should be substantiated here;

2.   Identify means to improve the utilization of the produced outputs and of the established systems and
     procedures (including corrective measures both internal and external to the project);

3.   Assess efficiency, effectiveness and impact, compared to similar projects, making concrete
     recommendation(s) on the course of actions to be followed.


                                                        18
4.   Extract policy lessons learned that will help to improve the design of other projects in the same
     sector/thematic area, particularly regarding UNCDF policy goals in LDF

5.   The policy lessons learned should include recommendations geared towards the sustainability of the
     LDF, and/or for improving the chances of similar projects to be up-scaled or replicated by the Government
     and other donors.




                                                     19
ANNEX II: LIST OF DOCUMENTS PRODUCED BY AIT.
List of Publications on Cambodia LDF
(As of March 1999)

   No.        Subject/Mission                    Title                        Author (s)        Date

    1.    LDF Formulation Report     Pilot Phase Design               Hiran.D. Dias
                                     Conceptual Background            H.Detlef. Kammeier   November 1995
                                    Papers                            Soparth Pongquan
                                                                      Paul Smoke
    2.    Mission Report             Data Collection Formats          Soparth P.           February 1996
          January 16-27, 1996
    3.     Back-to-office Report      Agreement after the LPS      Leonardo Romeo          March 1996
                                      Workshop, 19-21 February,    (UNCDF)
                                    1996
    4.     Training Workshop         Workshop Materials in English Soparth Pongquan        April 1996
                                                                   Hiran D. Dias
    5.     Workshop Report            Training of Trainers and     Soparth Pongquan        April 1996
                                      Support Staff                Hiran D. Dias
    6.     LDF Management                                          H. Detlef Kammeier      April 1996
           Procedures Interim                                      Paul Smoke
           Report
    7.     Standard Forms                                             H.Detlef Kammeier    April 1996
           “Templates”
    8.     Mission Report            First Monitoring (Pilot)         Soparth Pongquan     June 1996
           May 16-26, 1996
    9.     TSS Workshop Report       Training of Technical Support    H.Detlef Kammeier    July 1996
                                     Staff (TSS) on the Forms and     D.P. Villeneuve
                                     “Templates”
   10.     LPIS Report                Local Planning Information      H.D. Dias            July 1996
                                      System (LPIS)
   11.     Mission Report             Second Monitoring (Pilot)        Soparth Pongquan    July 1996
           June 22 – July 4, 1996
   12.     Mission Report             Third Monitoring                 Soparth Pongquan    August 1996
           July 23-30, 1996           “Evaluation of the First
                                       Pilot Communes”
   13.     Mission Report             Fourth Monitoring                Soparth Pongquan    October 1996
           October 3-12, 1996         “Evaluation of the Pilot
                                      Communes”
   14.     Training Workshop           A Compilation of LPP            Marilou P.Juanito   December 1996
             (Main Phase)              Training Materials              UNDP/CARERE
   15.     LDF Project                 Main Phase Design               Harvey Demaine      December 1996
           Formulation                                                 H.Detlef Kammeier
           Report (Main Phase)                                         Soparth Pongquan
                                                                       Paul Smoke
   16.     Mission Report      •     Back-to-Office Report           Soparth Pongquan      February 1997
           January 22-31, 1997 •     Design of Monitoring            Laura Kullenberg
           (M&E Design for LDF) And Evaluation System                    (UNCDF)
   17.     Mission Report         Evaluation of the                    Soparth Pongquan    June 1997
           March 23 – April, 1997 Implementation of the Pilot          Paul Smoke
           (Pilot Evaluation)     Phase of the UNCDF
                                  Local Development Fund

   18.     LDF Main Phase             “Project Implementation          UNDP/CARERE          June 1997
           Implementation              Manual” for LPP
   19.     M & E Report              Independent Monitoring           Robin Biddulph        July 1997
           (First Cycle of LPP       and Evaluation of the Local      Chor Vanna
            Main Phase)              Planning Process                 UNDP/CARERE


                                                          20
 No.        Subject/Mission                   Title                       Author (s)          Date


 20.       LDF/LPP Main             Monitoring of the Main Phase    Harvey Demaine       November 1997
           Phase Monitoring       Local                             Soparth Pongquan
           October 7-18, 1997        Planning Process               Paul Smoke
21.       M&E Report (LPP/LDF) Back-to-Office Report                Soparth Pongquan     January 1998
           January 6-9, 1998        on Monitoring and Evaluation
 22.       Financial System         Developing an                    Pa ul Smoke         February 1998
           January 4-15, 1998     Intergovernmental
                                    Fiscal System in Cambodia :
                                  Possible
                                    Models and Implications for
                                  the
                                    Local Development Fund
 23.     Linkage between PDP        The Integration of the Local      Harvey Demaine     April 1998
         and LPP                  and
         (February 18 – March 3, Provincial Planning Processes
          1998)                   and
                                    the Local Planning Information
                                    System
 24.     M&E Report                 Commune Development               Soparth Pongquan   April 1998
         March 17-28, 1998        Committee
                                    IPF Allocation Process
 25.     Financial System           Possible Structure of a           Paul Smoke         April 1998
         March 21-27, 1998        Provincial
                                    Development Fund
 26.     MTA CARERE 2               Mid-term Evaluation of           UNDP                 July 1998
         (April 27 - May 15, 1998) “CARERE 2”
 27.     MTA CARERE 2               Mid-term Evaluation of           Agnes Cargamento    August 1998
         (April 27 – May 15,        “CARERE 2” Local Planning      (UNCDF Consultant)
       1998)                        Process

 28.    M&E on LPP                  Independent Monitoring and       Robin Biddulph      August 1998
                                    Evaluation of CARERE                  Et al
                                 Support to                          (UNDP/CARERE)
                                    the SEILA Local Planning
                                 Process,
                                    1998
 29.     Review of the Second       Review of the Second Cycle       Harvey Demaine      October 1998
         Cycle of LPP/LDF        of                                  Leonardo Romeo
         (August 20 – September Local Planning Process (LPP)/ Soparth Pongquan
       3                            Local Development Fund           Paul Smoke
          1998)                  (LDF)
 30.     Basic Legislation         Assistance to the Ministry of     Patrick Roome       December 1998
         (December 12-25, 1998) Interior                           (UNCDF Consultant)
                                   on Basic Legislation for
                                 Decentralised
                                  Commune Councils
 31.     LPP Manual                Local Planning Process            UNDP/CARERE         January 1999
                                 Manual
                                              (Draft)
 32.     Finance Manual            The SEILA Finance Manual         UNDP/CARERE          February 1999
                                                                  (Mohammed Elmensi)
 33.     Decentralization Policy Back-to-Office Report on
         Statement and             Mission to Assist Ministry of    Soparth Pongquan      March 1999
       Commune                   Interior
         Government Law            in Inter-Ministerial Workshop,
         (February 10-26, 1999) Decentralization Policy
                                 Statement and
                                   Drafting Commune
                                 Government Law


                                                      21
ANNEX III ITINERARY
                        UNCDF EVALUATION MISSION August 22 – September 18
             Mr Bob Leonard [BL] (Team Leader), Mr Robert Guild [RG] (Infrastructure Specialist),
                            Ms Cristina Mansfield [CM] (Participation Specialist)


Date and Time       Activity                                   Location       Topic of Discussion

Monday, August 16 – Tuesday, August 17
                 Briefing with Bettina Fuhrmann, UNCDF         UNCDF, NYC     Briefing on LDF/Seila
                 Program Officer, Leonardo Romeo, Senior
                 Technical Advisor, UNCDF, Eddie Yee
                 Woo Guo, Evaluation Specialist

Saturday, August 21
1000 – 1200       Meeting with Dr. Soparth Pongquan, AIT       AIT, Bangkok   Pre-mission briefing and
                  UNCDF Desk Coordinator                                      project orientation

1200-1400           Lunch meeting with Dr. Soparth             AIT, Bangkok   Project constraints and
                    Pongquan, Dr. Detlef Kammeier, Dr.                        opportunities
                    Harvey Demaine

Monday, August 23
0800-0900           Meeting with Scott Leiper, Program         CARERE         Program overview
                    Manager, & Joel Charny, Deputy Program
                    Manager

0900-1000           Meeting with Provincial Project Managers   CARERE         Introduction to the mission

1030-1200           Meeting with H.E. Chhieng Yanara,          CDC            Review of the Seila Program
                    Cambodian Rehabilitation and                              and recent Royal Government
                    Development Board (CRDB) & Secretary                      Decrees related to the Seila
                    of the Seila Task Force                                   Task force and structure

1430-1600           Meeting with Jean-Claude Rogivue,          UNDP           Review of TOR and UNDP
                    Deputy Resident Representative, and                       Program for Administrative
                    Claire Van der Veer, Assistant Resident                   Reform
                    Representative


Tuesday, August 24
1030-1130        Meeting with HE Prum Sokha, Secretary         Ministry of    Discussion of Commune
                 of State of Interior, Sok Setha, Deputy       Interior       Administration Law, currently
                 Director of General Administration, and                      being drafted by Ministry of
                 Leng Vy, also Deputy Director of General                     Interior, and implications for
                 Administration.                                              Seila Program.

01400-1600          Meeting with Mohammed Elmensi,             CARERE         Seila financial systems, LDF,
                    Financial Systems Advisor                                 government finance




                                                         22
Wednesday, 25 August
0830-0930       Meeting with Mr. Yim Sam Oeun,               UNDP              Role of MRD in Seila Program
                Director, Department of Community
                Development, Ministry of Rural
                Development, Mr. Set Samoeun, Deputy
                Director, Office of Rural Economics, and
                Lang Mony, Deputy Director of
                Community Development

0930-1200          Review documentation                      UNDP

1400-1700          Briefing with Joel Charny, Deputy         CARERE            Question and answer session in
                   Program Manager                                             preparation for field trip


Date and Time      Activity                                  Location          Topic of Discussion

Thursday, August 26
0800-0900        Meeting with H.E. Sum Manit, Secretary      Council of        Public Administrative Reform
                 of State, Council of Ministers, and Mr.     Ministers         and implications for the Seila
                 Hongly Ngo, Program Manager, UNDP                             Program.
                 Support for Administrative Reform

1430               Departure for Battambang Province

1600               Meeting with CARERE PSO Staff             PSO               Overview of Battambang
                                                                               program and discussion of
                                                                               field visits

Friday, August 27
0830-1000         Meeting with HE Nov Sam, Battambang        PRDC              Review of mission objectives,
                  Governor, Mr. El Soy, Director,                              general discussion of Seila
                  Department of Rural Development and                          Program, PRDC presentation
                  Deputy Director of PRDC, and members                         of overview of Seila Program.
                  of the PRDC Executive Committee

1000-1200          RG meets with TSS and Finance Unit        PRDC              Technical Support, new Seila
                                                                               financial procedures,
                                                                               construction supervision

1000-1200          Concurrent meeting with PFTs and DFTs     PRDC              Challenges/successes of village
                                                                               and commune facilitation.


1400               Field visit to Reang Kesei Commune        School in Reang   Meeting with 20 members of
                   Sangke District                           Kesei Commune     CDC currently in its 3 rd cycle.
                                                                               Visit to infrastructure projects
                                                                               in commune.




                                                        23
Saturday, August 28
0800              Departure for Sampeuv Luon District                             Visit reconciliation district

1400-1600           Meeting with Phon Phen, First Deputy        District Office   Overview of reconciliation
                    District Chief of Sampeuv Luon and 20                         process, priorities for district
                    members of DDC                                                development, LPP

1600-1700           Meeting with Ta Sra Commune CDC             House of Ta Sra   Commune Development Plan
                                                                Commune Chief

Saturday, August 28
0800-             RG inspects projects with Chan Sambo,                           Inspection of Infrastructure
                  TSS Staff and Tha Thai, DFT member:                             Projects

                    Projects visited in Preah Punea with VDC    Rung Chray
                    Chief Sim Kong and CDC Chief Oy Siem        Commune,
                    Rung. In Pouy Rumchey, projects visited     Banan District
                    with VDC Chief Chen Chun and Village
                    Chief Chun Chee. Projects also visited in
                    Balang Kron and Prokeap villages.           Rung Chray
                                                                Commune,
                    Projects visited in Kok Ream with San Ok,   Banan District
                    VDC Chief, and Srey Ngoun, Deputy CDC
                    Chief. In Knack Romeas Village, projects
                    visited with Lang Choi, VDC Chief.

Date and Time       Activity                                    Location          Topic of Discussion

Sunday, August 29
0800              Departure for Phnom Prik District

1000-1200           Meeting with Phoeun Napheng, Phnom          Temporary         Discussion of LPP with second
                    Prik District Chief and District            District Office   reconciliation district
                    Development Committee

1400                Return to Battambang Town

Sunday, August 29
0800-1700         RG visits projects in Phnom Sampou            Phnom Sampou      Inspection of Infrastructure
                  Village with Commune Chief Ou Loun            Commune,          Projects
                  and CDC and VDC members.                      Sangke District

                    Visit to Svray Prakeap Village.             Chouteal
                                                                Commune




                                                        24
Monday, August 30
0800-1200        Mansfield on field trip to Prek Norin        Commune
                 Commune                                      Office              Discussion of
                                                                                  LPP, LDF
0800-0900           Meeting with Mr. Tiev Chulong, Deputy     PRDC
                    Director, Department of Planning                              allocation

0900-1015           Meeting with Mr. Niv Narin, Director,     PRDC
                    Department of Agriculture

1015-1130           Meeting with Mr. El Soy, Director,        PRDC                Data collection
                    Department of Rural Development

1400-1500           Meeting with Johann Siffointe, UNHCR      UNHCR

1500-1600           Meeting with Mr. Neang Hy, Assistant      CIHR
                    Local Manager, Cambodian Institute for                        Agriculture
                    Human Rights (CIHR)                                           project
1600-1630           Meeting with Mr. Im Luon, Director,       HURIPRUDA
                    Human Rights Protection and
                    Development Association (HURIPRUDA)

1630-1700           Meeting with Khem Yari, CD                LWS
                    Officer, and Sam Tonleng,                                     Road and
                    Agriculture Advisor, Lutheran                                 watsan projects
1730-1800           World Service (LWS)                       ANS

                    Meeting with Mr. Som Sokha,
                    Coordinator, Local Development Agents,                Reconciliation program
                    Action Nord Sud (ANS)
                                                                          Human rights training for
                                                                          VDCs


                                                                          Seila Non-Infrastructure
                                                                          Projects implemented by
                                                                          HURIPRUDA

                                                                          Collaboration with Seila



                                                                          Collaboration with Seila


Tuesday, August 31
0800-1200        Preparation for debriefing with PRDC         Hotel

1500-1700           Meeting with Battambang PRDC and          PRDC        Debriefing of team findings
                    CARERE staff                                          and PRDC response




                                                         25
Date and Time       Activity                                      Location         Topic of Discussion

Wednesday September 1
1000            Arrive Sisophon                                   Sisophon

1000-1100           Meeting Peter Robertson (PPM), Kong           PSO              Introduction, review itinerary,
                    Sokuntho (DPPM)                                                handover documents

1100-1200           Meeting CARERE PSO staff – LCBs,              PSO              Introductions discussion LPP
                    M&E, Infra. Assistants

1430-1600           Meeting with H.E. Thach Khon                  PRDC             Introductions, briefing from
                    (Governor), PRDC Ex. Com.                                      PRDC SEILA/LDF/LPP

1600-1730           Meeting Secretariat staff PFT/DFT/            PRDC             RG discuss with TSS
                    TSS/Finance/M&E/ Resource Center.
                    Ros Sophon (Sec Chief, Dir. Dept Plg)                          BL/CM meet with PFT/DFT
                                                                                   and M&E Unit
Thursday September 2
0700              Depart Sisophon for Preah Netr Preah            PNP District     District Integration workshop
                                                                  Office
                    RG field visits to LDF projects in                             RG - Visit LDF projects
                    Prasat Commune
1800
                   Arrive in Sisophon
Friday September 3
0730               Depart Sisophon for Koh Pong Sat               CDC office       Discussion with CDC on
                   (Sereisophon District)                                          LDF/LPP

1400                Depart for Slor Kram CDC (BL)                 CDC office       Discussion with CDC from
                                                                                   UNCDF pilot commune on
                                                                                   LDF/LPP

1400                RG meets with Director of Dept of             Sisophon         SEILA Finance System
                    Finance/Finance Unit
1700                                                                               Gender mainstreaming

                    CM Director of Dept
1800                Womens/Veterans Affairs

                   BL arrives in Sisophon
Saturday September 4
0700               Depart Sisophon for Malai

0900                Attend O’Sampor Bidding meeting               O’Sampor         Attend Bidding meeting
                                                                  school

1430-1600           Meeting with Malai District Governor          Malai District   Discussion of reconciliation
                    (Chhim Buny) and other district authorities   office           programme and SEILA




                                                         26
Sunday September 5
0730               Depart for Koy Meng by car then boat to       Dong Treng        Discuss with CDC from PnP
                   Dong Treng Village (Preah Netr Preah                            and Koy Meng and with Dong
                   District)                                                       Treng villagers about a project
                                                                                   which doesn’t fit within LDF
1200-1500           Picnic lunch                                 Dong Treng        resource allocation

Date and Time     Activity                                       Location          Topic of Discussion
Monday September 6
0600              Depart for Banteay Chhmar/Kok O’lok

0900                VDC/CDC training on roles and                Kok O’lok         Discussion with new VDC/
                    responsibilities –BL/CM                      school            CDC in reconciliation areas

                    RG visit LDF projects in Kok                 Kok Romiet        Visit reservoir project in
                    Romiet Commune (Kdip Thmar)                  Kdip Thmar        reconciliation village

1400                                                             Banteay Chhmar    Discussion with new VDC/
                    VDC/CDC training on roles and                school            CDC in reconciliation areas
                    responsibilities – BL/CM

                    RG visit projects
1830
                    Arrive in Sisophon

Tuesday, September 7
0800            Depart for Siem Reap by car

1500                Briefing with Hans van Zoggel, Provincial    PSO               Presentation of Siem Reap
                    Program Manager, Duong Vanna, Deputy                           Program
                    Provincial Program Manager, and Julian
                    Abrams, Infrastructure Advisor

Wednesday September 8
0730                CM to district integration workshop in       District Office   Observe district integration
                    Soutr Nikom District accompanied by                            workshop (with participation
                    Duong Vanna and Soutr Nikom LCB,                               of non-target communes)
                    meeting with VDC Chief.

                    RG on field visit and system analysis with
                    Julian Abrams, Infrastructure Advisor

1400                CM to district integration workshop in Chi                     Observe district integration
                    Kraing                                                         workshop (with participation
                                                                                   of non-target communes)




Thursday, September 9
0730-1200        Orientation meeting for newly-formed            Treynhor          Newly-formed/experimental
                 experimental VDC in Theas Village               Commune,          VDCs
                 accompanied by Nhim Hak, LCB.                   Phuok District

0730-1500           RG on field visits to Sautr Nikom District   Sautr Nikom       Visit infrastructure projects
                    to inspect infrastructure projects           District

                                                         27
Friday, September 10
0800              Meeting with Siem Reap Provincial          PSO          Scheduling, collecting
                  Program Manager                                         information

0930               CM meets with Deputy Director of PDRD     PDRD         Provincial and district
                                                                          facilitation

1500               RG meets with Finance Director                         P
                                                                          R
                   BL meets with Planning Director                        D
                                                                          C
1600               Debriefing with Excom                                  Debriefing

                                                             PRDC

                                                             PRDC

Date and Time      Activity                                  Location     Topic of Discussion

Saturday, September 11
0900-1200        Debriefing with CARERE Provincial           PSO          Debriefing
                 Program Manager

Sunday, September 12
                Team Returns to Phnom Penh

1800-2100          RG meets with Lee Forsythe,               SRP          Databases, M&E
                   Reconciliation Program Coordinator


Monday, September 13
1100-1200       Meeting with Joel Charny, CARERE             CARERE
                Deputy Program Manager

1600-1700          Meeting with Someth Suos, Resident        ADB          ADB country programs
                   Representative and Nida Ouk, Projects
                   Officer, Asia Development Bank

1700-1800          Meeting with Bonaventure Mbida-Essana,    World Bank   WB country programs
                   Acting Chief, World Bank Resident
                   Mission

Tuesday, September 14
0900-1000       Meeting with Mr. Frederic Urfer, TA          PRASAC/EU    PRASAC rural development
                Coordinator, PRASAC Project, European                     model
                Union

Wednesday, September 15
0900            Meeting with Toshi Kato, Economist,          CDRI         Poverty Indicators
                Cambodian Development Research
                Institute (CDRI)



                                                        28
Thursday, September 16
1030-1200        Mission debriefing with UNCDF/UNDP   UNDP

Friday, September 17




                                                29
ANNEX IV               LIST OF PEOPLE INTERVIEWED

Last        First            TITLE                         DEPARTMENT                     AGENCY                 TOPIC

   NEW YORK

Fuhrmann    Bettina    Ms.   Program Officer                                              UNCDF                  Mission Briefing
Romeo       Leonardo   Mr.   Senior Technical Officer                                     UNCDF                  Mission Briefing
Yee Woo     Eddie      Mr.   Evaluation Specialist                                        UNCDF                  Mission Briefing
Guo


  BANGKOK

Demaine     Harvey     Dr.                                                                AIT                    Briefing
Kammeier    Detlef     Dr.                                                                AIT                    Briefing
Pongquan    Soparth    Dr.   AIT-UNCDF Desk                                               AIT                    Briefing
                             Coordinator



PHNOM PENH

Charny      Joel       Mr.   Deputy Program Manager                                       CARERE                 additional briefing
Chhieng     Yanara     Mr.   Secretary General             Cambodian Rehabilitation and   Council for            Seila Task Force
                                                           Development Board (CRDB)       Development of
                                                                                          Cambodia
Duong       Vanna      Mr.   Deputy Provincial Program     Siem Reap PSO                  CARERE                 General Orientation
                             Manager
Elmensi     Mohamed Mr.      Financal Systems Advisor                                     CARERE                 Seila, LDF, Gov Financial
                                                                                                                 Systems
Gross       Laurie     Ms.   UNCDF Representative                                         UNLDF                  Introduction
Hongly      Ngo        Mr.   Program Manager                                              UNDP                   Public Administration
                                                                                                                 Reform
Jan         Sophal     Mr.   Provincial Program            Pursat PSO                     CARERE                 General Orientation
            Chhor            Manager
Lang        Mony       Mr.   Deputy Director               Department of Community        Ministry of Rural      Role of MRD in Seila
                                                           Development                    Development
Leiper      Scott      Mr.   Program Manager                                              CARERE                 Introduction
Leng        Vy         Mr.   Deputy Director               General Administration         Ministry of Interior   Commune Administration
                                                                                                                 Law
Min         Muny       Mr.   Provincial Program            Ratanakiri PSO                 CARERE                 General Orientation
                             Manager
Morrison    Joanne     Ms.   Provincial Program            Battambang PSO                 CARERE                 General Orientation
                             Manager
Pietersen   Irene      Ms.   M&E Advisor                   Battambang/Banteay             CARERE                 General Orientation
                                                           Meanchey PSOs
Prum        Sokha      Mr.   Secretary of State                                           Ministry of Interior   Commune Administration
                                                                                                                 Law
Robertson   Peter      Mr.   Provincial Program          Banteay Meanchey PSO             CARERE                 General Orientation
                             Manager
Rogivue     Jean-      Mr.   Deputy Resident Representative                               UNDP                   Review of TOR
            Claude
Roome       Patrick    Mr.   Commune Law Advisor to MOI                                   UNDP                   Commune Administration
                                                                                                                 Law
Set         Samoeurn Mr.     Deputy Director, Office of Rural    Department of            Ministry of Rural      Role of MRD in Seila
                             Economics                           Community                Development
                                                                 Development
Sok         Setha      Mr.   Deputy Director General             General Administration   Ministry of Interior   Commune Administration
                                                                                                                 Law
Sum         Manit      Mr.   Secretary of State                                           Council of Ministers   Public Administration
                                                                                                                 Reform

                                                                  30
                                                                                                              Reform
Van der   Claire      Ms.   Assistant Resident Representative                            UNDP                 Review of TOR
Vaeren
van       Hans        Mr.   Provincial Program Manager          Siem Reap PSO            CARERE               General Orientation
Zoggel
Yim       Sam         Mr.   Director                            Department of            Ministry of Rural    Role of MRD in Seila
          Oeun                                                  Community                Development
                                                                Development


BATTAMBANG
Brak      Phy Run     Mr.   VDC Member                          Phnom Sampou Commune, Banan District,         Infrastructure Projects
                                                                Battambang
Chan      Sarath      Mr.   Deputy Director, Agriculture        Sampov Loeun District, Battambang             Reconciliation Program
Chay      Mao         Mr.   Commune Chief, Barang Tleak         Phnom Prik District, Temporary District       Reconciliation Program
                                                                Battambang           Office
Chea      Ny          Mr.   VDC Chief                           Reang Kaisei Commune Sangke,                  CDC workings future
                                                                                     Battambang               outlook
Chea      Sambath     Mr.   Deputy Chief of Secretariat         PRDC Secretariat     PRDC Battambang          PFT/DFT/M&E
Chen      Chun        Mr.   VDC Chief, Pouy Rumchey Village Rung Chray Commune, Banan District,               Infrastructure Projects
                                                            Battambang
Chhay     Chhoeun     Mr.   DFT, Banan                      PRDC Secretariat     PRDC Battambang              PFT/DFT/M&E
Chop      Saroeun     Mr.   Finance officer                     PRDC, Battambang                              Reconciliation Program
Chounh    Sochay      Mrs. Director                             Department of Women's Battambang              Debriefing with PRDC
                                                                Affairs
Choup     Jareoung    Mr.   --                                  Finance Unit          PRDC Secretariat        Financial management
Chun      Chee        Mr.   Village Chief, Pouy Rumchey         Rung Chray Commune, Banan District,           Infrastructure Projects
                            Village                             Battambang
Do        Phu         Mr.   Second Deputy Chief                 Sampov Loeun District, Battambang             Reconciliation Program
Doung     Vanna       Ms.   CDC Member, Gender focal point      Phnom Sampou Commune, Banan District,         Infrastructure Projects
                                                                Battambang
El        Soy         Mr.   Deputy Chair, PDRC                  Director, Dept Rural Battambang               Comments on PRDC,
                                                                Development          Province                 LPP, Seila
He Sa     Haip        Mr.   Deputy Chair, PDRC                  1st Vice Governor    Battambang               Comments on PRDC,
                                                                                     Province                 LPP, Seila
Hong      Chring      Mr.   Director                            Finance Unit         PRDC Secretariat         Financial management
Im        Luon        Mr.   Director                            Human Rights             HURPRUDA             Collaboration with Seila
                                                                Protection Development
                                                                Association
Johann    Siffointe   Mr.   Head of Office                                               UNHCR                Reconciliationn Program
                                                                                         Battambang
Kak       Ravy        Mr.   Office Chief, Community             Department of Rural      Battambang           Allocation, financial
                            Development                         Development                                   management, rural
                                                                                                              planning structure
Keang     Tinh        Mr.   CDC                                 Prek Luong Commune       Battambang
Keo       Rotha       Mr.   VDC Chief, O'Proyut                 Phnom Prik District,     Temporary District   Reconciliation Program
                                                                Battambang               Office
Khem      Yarin       Mr.   Community Development Officer       Lutheran World Service   LWS                  Collaboration with Seila
Kim       Srien       Ms.   Deputy District Officer,            Phnom Prik District,     Temporary District   Reconciliation Program
                            Department of Women' s Affairs      Battambang               Office
Kol       Ly          Mr.   M&E                                 PRDC Secretariat         PRDC Battambang      PFT/DFT/M&E
Kong      Som         Mr.   VA Chief                            Reang Kaisei Commune Sangke,                  CDC workings future
                                                                                     Battambang               outlook
Kung      Munichan Mr.      Deputy Provincial Program           Battambang PSO       CARERE                   Comments on PRDC,
                            Manager                                                                           LPP, Seila
Ky        Jerey       Mr.   --                                  Finance Unit             PRDC Secretariat     Financial management
Lach      Thea        Mr.   VDC Member                          Reang Kaisei Commune Sangke,                  CDC workings future
                                                                                     Battambang               outlook
Lang      Choi        Mr.  VDC Chief, Knack Romean              Knack Romeas Commune, Banan District,         Infrastructure Projects
                           Village                              Battambang
Long      Mong        Mrs. VDC Chief                            Reang Kaisei Commune Sangke,Battambang        CDC workings future
                                                                                                              outlook
Luoth     Phong       Mr.   Deputy Director                     Department of            Battambang           Debriefing with PRDC
                                                                Agriculture
Mai       Yong        Mr.   Deputy Commune Chief, Preak         Ek Phnom District,       Commune Office       LDF Allocation
                            Norin                               Battambang


                                                                31
Mang        Chan Tho Mr.    Resource Center                       PRDC Secretariat         PRDC Battambang      PFT/DFT/M&E
Mao         Sofhanna Mr.    Office Chief, Rural Water             Department of Rural      Battambang           Allocation, financial
                                                                  Development                                   management, rural
                                                                                                                planning structure
Me          Mai Sa    Mr.   Director, Hosp, Dep Dir Health        Sampov Loeun District, Battambang             Reconciliation Program
Me          Kom       Mr.   Police Inspector                      Sampov Loeun District, Battambang             Reconciliation Program
Mhang       Pou       Mr.   VDC Chief                             Reang Kaisei Commune Sangke,                  CDC workings future
                                                                                       Battambang               outlook
Mom         Bunna     Mr.   Local Planning Unit                   PRDC Secretariat     PRDC Battambang          PFT/DFT/M&E
Morrison    Joanne    Ms.   Provincial Program Manager            Battambang PSO           CARERE               Comments on PRDC,
                                                                                                                LPP, Seila
Mou         Sarin     Mr.   DFT, Sampeou Loeun                    Sampov Loeun District, Battambang             Reconciliation Program
Nay         Chiroh    Mr.   Village Chief, O'Lahong               Phnom Prik District,     Temporary District   Reconciliation Program
                                                                  Battambang               Office
Neang       Hy        Mr.   Assistant Local Manager                                        CIHR Battambang      HR training for Seila
                                                                                           Community Center     Communes
Net         Nat       Mr.   Deputy Director, Religious Affairs    Sampov Loeun District, Battambang             Reconciliation Program
Niv         Narin     Mr.   Director                              Department of            Battambang           Integration process
                                                                  Agriculture
Nov         Sam       H.E. Governor and PRDC Chairman                                      Battambang           Debriefing with PRDC
Ou          Loun      Mr.   CDC Chief                             Phnom Sampou Commune, Banan District,         Infrastructure Projects
                                                                  Battambang
Oui         Chan      Mr.   Village Chief, Samrong                Phnom Prik District, Temporary District       Reconciliation Program
                                                                  Battambang           Office
Ouk         Pun       Mr.   DFT, Svay Por                         PRDC Secretariat     PRDC Battambang          PFT/DFT/M&E
Oum         Sam On    Ms.   PFT, Banan                            PRDC Secretariat         PRDC Battambang      PFT/DFT/M&E
Oy          Siem      Mr.   CDC Chief                             Rung Chray Commune, Banan District,           Infrastructure Projects
            Rung                                                  Battambang
Pan         Keu       Mr.   VDC Chief, Samhang Village            Phnom Sampou Commune, Banan District,         Infrastructure Projects
                                                                  Battambang
Pan         Tum       Mr.   Commune Chief, Rumdoul                Phnom Prik District,     Temporary District   Reconciliation Program
                            Commune                               Battambang               Office
Pheun       Napheng   Mr.   District Chief                        Phnom Prik District,     Temporary District   Reconciliation Program
                                                                  Battambang               Office
Phon        Phen      Mr.   First Deputy Chief                    Sampov Loeun District, Battambang             Reconciliation Program
Phuok       Vuthy     Mr.   LCB Team Leader                                                CARERE               PFT/DFT/M&E
Prak        Nun       Mr.   Village Chief,                        Phnom Prik District,     Temporary District   Reconciliation Program
                                                                  Battambang               Office
Represent                   Office of Statistics                  Sampov Loeun District, Battambang             Reconciliation Program
ative
Reung       Nan       Mr.   VDC Chief, Sampou Village             Phnom Sampou Commune, Banan District,         Infrastructure Projects
                                                                  Battambang
Rin         Rany      Mr.   VDC Member, Kandal Village            Phnom Prik District,   Temporary District     Reconciliation Program
                                                                  Battambang             Office
Sam         Tonleng   Mr.   Agriculture Advisor                   Lutheran World Service LWS                    Collaboration with Seila
San         Senghak   Mrs. VDC Member                             Reang Kaisei Commune Sangke,                  CDC workings future
                                                                                           Battambang           outlook
San         Ok        Mr.   VDC Chief, Kok Ream Village           Knack Romeas Commune, Banan District,         Infrastructure Projects
                                                                  Battambang
Sao         Sary      Mr.   Deputy Director, Education            Sampov Loeun District, Battambang             Reconciliation Program
Sat         Soeung    Mr.   CDC                                   Prek Luong Commune       Battambang
Say         Rim       Mr.   Deputy Director, District Office of   Phnom Prik District,     Temporary District   Reconciliation Program
                            Education                             Battambang               Office
Seng        Valath    Mr.   Chief, PRDC Secretariat               Director, Planning       Battambang           Comments on PRDC,
                                                                                           Province             LPP, Seila
Seng        Mary      Ms.   Deputy Director                       Department of Women'     Battambang           Comments on PRDC,
                                                                  s Affairs                Province             LPP, Seila
Seng        Yi        Mr.   Commune Chief, Preak Norin            Ek Phnom District,       Commune Office       LDF Allocation
                            Commune                               Battambang
Seth        Bony      Mr.   M&E                                   PRDC Secretariat         PRDC Battambang      PFT/DFT/M&E
Sim         Kong      Mr.   VDC Chief, Preah Punea Village        Rung Chray Commune, Banan District,           Infrastructure Projects
                                                                  Battambang
Siv         Seneh     Ms.   PFT, Sangke                           PRDC Secretariat     PRDC Battambang          PFT/DFT/M&E
So          Sovath    Mr.   Deputy Director                       Department of Rural      Battambang           Allocation, financial
                                                                  Development                                   management, rural
                                                                                                                planning structure

                                                                  32
                                                                                                       planning structure
Sok       Soeur      Mr.   VDC Member                       Reang Kaisei Commune Sangke,               CDC workings
                                                                                     Battambang
Sok       Kroy       Mr.   Director, Land Title Office      Sampov Loeun District, Battambang          Reconciliation Program
Som       Reth       Mr.   Commune Chief, CDC Chief         Reang Kaisei Commune Sangke,               CDC workings
                                                                                 Battambang
Som       Vanna      Mr.   CDC Secretary                    Reang Kaisei Commune Sangke,               CDC workings
                                                                                 Battambang
Som       Saron      Mr.   VA Chief                         Reang Kaisei Commune Sangke,               CDC workings
                                                                                 Battambang
Som       Sokha      Mr.   Coordinator, Local Development   Action Nord Sud      ANS                   Collaboration with Seila
                           Agent
Srey      Ngoun      Mr.   Deputy VDC Chief, Kok Ream       Knack Romeas Commune, Banan District,      Infrastructure Projects
                           Village                          Battambang
Tach      Savy       Mr.   Finance Officer                                      CARERE,                Financial management,
                                                                                Battamang PSO          new Seila structure
Tat       Ny         Mr.   Deputy Director                  PRDC Secretariat    Battambang             Debriefing with PRDC
Teav      Choulong Mr.     Deputy Director                  Department of Rural  Battambang            Debriefing with PRDC
                                                            Development
Tep       Rany       Mrs. CDC Member                        Reang Kaisei Commune Sangke,               CDC workings future
                                                                                 Battambang            outlook
Tha       Thavy      Mr.   DFT, Battambang                  PRDC Secretariat     PRDC Battambang       PFT/DFT/M&E
Theam     Sam El     Mr.   TSS, Sangke                      PRDC Secretariat        PRDC Battambang    CDC workings future
                                                                                                       outlook
Thoung    Chhem      Mr.   CDC Member                       Phnom Sampou Commune, Banan District,      Infrastructure Projects
                                                            Battambang
Tiev      Choulong Mr.     Deputy Director                  Department of Planning Battambang          LPP, monitoring, data, IPF
                                                                                                       allocation
Touch     Samnang Mr.      DFT, Sangke                      PRDC Secretariat        PRDC Battambang    CDC workings future
                                                                                                       outlook


       BANTEAY
      MEANCHEY

Bun       An         Mr.   CDC Chief                        Preah Pongsat           Banteay Meanchey   Impact, contributions,
                                                            Commune, Serey                             allocation, district
                                                            Sophon District                            integration, planning
                                                                                                       process
Chim      Kriphong   Mr.   Director, Finance Department     PRDC                    Banteay Meanchey   Seila financial procedures
Chim      Mony       Mr.   District Chief                   Malai District          Banteay Meanchey   Reconciliation Process,
                                                                                                       LPP
Chong     Phet       Mr.   District Chief                   Preah Netr Preah        Banteay Meanchey   District Integration
                                                            District                                   Workshop
Heang     Heang      Mr.   VDC Chief, Preah Ang             Preah Pongsat           Banteay Meanchey   Impact, contributions,
                                                            Commune, Serey                             allocation, district
                                                            Sophon District                            integration, planning
                                                                                                       process
Ith       Loeur      Mr.   Director                         PRDC Secretariat        Banteay Meanchey   Introduction of UNCDF
                                                                                                       Team
Keo       Ngim       Mr.   Village Chief, Koh Pongsat       Preah Pongsat           Banteay Meanchey   Impact, contributions,
                                                            Commune, Serey                             allocation, district
                                                            Sophon District                            integration, planning
                                                                                                       process
Kun       Koy        Mr.   CDC Deputy Chief                 Preah Pongsat           Banteay Meanchey   Impact, contributions,
                                                            Commune, Serey                             allocation, district
                                                            Sophon District                            integration, planning
                                                                                                       process
Leak      Tong       Mr.   Deputy Director                  PRDC                    Banteay Meanchey   Meeting with PFT/DFT
Leang     Chon       Mr.   PFT                              Preah Pongsat           Banteay Meanchey   Impact, contributions,
                                                            Commune, Serey                             allocation, district
                                                            Sophon District                            integration, planning
                                                                                                       process
Lek       Cheo       Mr.   CDC Member                       Preah Pongsat           Banteay Meanchey   Impact, contributions,
                                                            Commune, Serey                             allocation, district
                                                            Sophon District                            integration, planning
                                                                                                       process
Long      Silux      Ms.   Senior LCB                                               CARERE BMC         Meeting with PFT/DFT


                                                            33
Meas       Iat        Mr.   VDC Chief, Snay Dangkor             Preah Pongsat            Banteay Meanchey   Impact, contributions,
                                                                Commune, Serey                              allocation, district
                                                                Sophon District                             integration, planning
                                                                                                            process
Nget       Muon       Mr.   Director, Office of Rural           Malai District           Banteay Meanchey   Reconciliation Process,
                            Development                                                                     LPP
Nhek       Sara       Ms.   Director, Department of Women's                              Banteay Meanchey   Gender Issues
                            Affairs
Phi        Phuon      Mr.   Deputy District Chief               Malai District           Banteay Meanchey   Reconciliation Process,
                                                                                                            LPP
Rath       Vanna      Ms.   Accountant                          PRDC                     Banteay Meanchey   Seila financial procedures
Ros        Sophon     Mr.   Director (Sec Gen of PRDC)          Department of Planning   Banteay Meanchey   Introduction of UNCDF
                                                                                                            Team
Si         Ron        Mr.   VDC Chief, Ankeal Boh Village       Preah Pongsat            Banteay Meanchey   Impact, contributions,
                                                                Commune, Serey                              allocation, district
                                                                Sophon District                             integration, planning
                                                                                                            process
So         Samoeun Mr.      VDC Chief, Ko Pongsat               Preah Pongsat            Banteay Meanchey   Impact, contributions,
                                                                Commune, Serey                              allocation, district
                                                                Sophon District                             integration, planning
                                                                                                            process
Sum        Socheath   Mr.   Director, PRDC Finance Unit         PRDC                     Banteay Meanchey   Seila financial procedures
Thach      Khon       H.E. Governor                                                      Banteay Meanchey   Introduction of UNCDF
                                                                                                            Team
Thip       Samay      Mr.   DFT                                 Preah Pongsat            Banteay Meanchey   Impact, contributions,
                                                                Commune, Serey                              allocation, district
                                                                Sophon District                             integration, planning
                                                                                                            process
Um         On         Mr.   CDC Member                          Preah Pongsat            Banteay Meanchey   Impact, contributions,
                                                                Commune, Serey                              allocation, district
                                                                Sophon District                             integration, planning
                                                                                                            process


      SIEM REAP

Abrams     Julian     Mr.   Infrastructure Advisor                                       CARERE SR          Briefing
Bou        Vanna      Mr.   Watsan and Infrastructure Advisor                            CARERE SR
Chan       Sophal     Mr.   Director                            Finance Unit             PRDC               Seila financial procedures
Chhun      Dy         Mr.   Watsan Advisor                                               CARERE SR
Duong      Vanna      Mr.   Deputy Provincial Project                                    CARERE SR          Briefing
                            Manager
Em         Phallamoy Ms.    Deputy Director, Department of Women's Affairs               Banteay Meanchey   District Integration
Forsythe   Lee        Mr.   Coordinator                         Reconciliation Programs CARERE              M&E
Kao        Sophon     Mr.   District Chief                      Chi Kraing District      Banteay Meanchey   District Integration
Keochan    Rotha      Mr.   Deputy Director, Department of Rural Development             Banteay Meanchey   District Integration
Khut       Douen      Mr.   Contractor                                                                      bidding procedures
Kim        Sia        Mr.   Directory, Department of                                     Banteay Meanchey   District Integration
                            Hydrology
Kim        Bon Nung Mr.     Technical Support Staff                                      PRDC               bidding procedures
Kong       Sothak     Mr.   Infrastructure Assistant                                     CARERE SR          Briefing
Kong       Ty         Mr.   Technical Support Staff                                      PRDC               bidding procedures
Kong       Soparth    Mr.   Technical Support Staff                                      PRDC               bidding procedures
La         Oum        Ms.   Department of Women's Affairs                                Banteay Meanchey   District Integration
Ly         Hey        Mr.   Contractor                                                                      bidding procedures
Ma         Set        Mr.   Technical Support Staff                                      PRDC               bidding procedures
Morn       Chun       Mr.   Contractor                                                                      bidding procedures
Ros        Roueng     Mr.   Contractor                                                                      bidding procedures
Seng       Vanna      Mr.   Contractor                                                                      bidding procedures
Soat       Pisak      Mr.   Director                            Department of Planning, PRDC                PRDC structure
                                                                Siem Reap
Than       Darravuth Mr.    Technical Support Staff                                     PRDC                bidding procedures



                                                                34
van       Hans    Mr.   Provincial Project Manager        CARERE SR   Briefing
Zoggell
Vun       Rotha   Mr.   Contractor                                    bidding procedures




                                                     35
ANNEX V : THE RURAL DEVELOPMENT STRUCTURE


Seila Task Force


Provincial Rural Development Committee (PRDC)
The PRDC was established by sub-decree January 1995 as the official CARERE2 counterpart. Its
primary role is to effectively manage the provincial development process for which it will receive
technical and financial assistance from UNDP/CARERE2. The responsibilities of the PRDC extend
beyond the CARERE2 project to any government or agency programs. It is expected that the PRDC
will increasingly become the focal point for coordination of most of the provincial development
activities. The PRDC is Chaired by the Provincial Governor and its members include the directors of
the provincial sector departments, as well as well as the provincial police commissioner, district chiefs,
and the provincial CMAC Manager. In addition, representatives of UN, international organizations,
and non-governmental organizations are admitted as observers to the PRDC.

PRDC Executive Committee (EXCOM)
The PRDC EXCOM is chaired by the Governor and includes Directors of the Departments of Rural
Development, Planning, Economy and Finance, Women’s Affairs, and Agriculture, and the Chief of
the EXCOM Secretariat. The EXCOM is charged with managing the execution of the Seila Investment
Plan (SIP) and other development agreements that fall within their mandate. Among other tasks, the
EXCOM is responsible for managing and being accountable for resources provided through the SIP,
appraising the SIP prepared by the Department of Planning, appraise contract proposals submitted by
designated implementing agencies and recommending them for approval y the Chair of the EXCOM,
and producing regular progress reports for the Task Force.

District Development Committee (DDC)
The DDC is chaired by the district chief and includes directors of the district technical offices, the
district police inspector, the chairpersons of the CDCs, and a designated district facilitator. The overall
responsibilities of the DDC are to facilitate the flow of information to the CDCs and VDCs, provide
technical support and training, prepare the district work plan describing the resources of the various
implementing agencies engaged in development in the district, facilitate the work of the provincial
technical departments in the district, establish a resource center to manage information and keep
records, and ensure that the appropriate and participatory monitoring of development activities within
the district is carried out.

Commune Development Committee (CDC)
The CDC is made up of the Chief and Deputy Chief of the Commune (appointed by the Royal
Government), representatives of the district technical offices, chairpersons of the VDCs, elected
women representatives, a commune assistant, and other local leaders (including the village chiefs),
who act as advisors. The primary role of the CDC is to serve as a local management body to
effectively manage the local development process within the commune, ensuring the active
participation of all villagers. The Commune Chief is the Chairman of the CDC.




                                                     36
Village Development Committees (VDC)
The local planning process begins at the village level with the election of VDCs made up of 5 to 7
members. The elections are organized by the PDRD. As an elected village body, the VDC is
recognized by the Royal Government as an autonomous body that will work to ensure coordination
and communication between local authorities and civil authorities. The VDC in turn recognizes the
authority of the Government represented by the village chief. Under the Seila Program, the primary
role of the VDC is to fully ensure the full participation of the village in the development process.

Elected VDC members are expected to hold their position for three years, after which new elections
are held. In the event that a VDC member resigns, another election is held to choose a replacement.
Each VDC is recognized by the PDRD through an official letter and the PDRD issues guidelines on the
task of each VDC member. The officers of the VDC are the Chief, Deputy Chief, Finance Officer,
Information Officer and the Gender Officer.




                                                   37
Annex VI    RGC Sub-Decree
                            KINGDOM OF CAMBODIA
                            NATION RELIGION KING
Royal Government
No 78ANKR.BK

                                 Sub-Decree
                    on Establishment of SEILA Task Force
                              The Royal Government
- Noting the Constitution of the Kingdom of Cambodia.
- Noting the Royal Decree No SN/RKT/1198/72 dated 30 November 1998 on
  Appointment of the Royal Government.
- Noting the Royal Decree No 02/SN dated 20 July 1994 promulgating the
  establishment and operation of the Council of Ministers.
- Noting the Sub-decree No 20 ANKR.BK dated 30 April 1996 on the establishment
  and operation of the Ministries and State's secretariat.
- Based on the draft proposed by Senior Minister, Minister of Economy and Finance

                           IT IS HEREBY DECIDED

                                     Chapter 1
                                  General Provision

Article 1.- The sub-decree defines the establishment of a Task Force at national level to
           manage the decentralized development programme of the Royal Government
           entitled the SEILA Programme with collaboration implemented by the Rural
           Development Committees at all levels.

Article 2.- The goals of the SEILA Programme are:

- to contribute to poverty alleviation in rural areas through implementation of
  decentralization policy;
                                           38
- to design and operationalize decentralized systems of planning, finance and
  management of provincial development;
- to evaluate the effectiveness of the decentralized systems' contribution to the
  implementation of national goals.




                                       39
                                        Chapter 2
                                       Composition

Article 3.- Membership of the SEILA Task Force is as follows:
       1. Minister of Economy and Finance                                    Chairman
       2.       Secretary of State, Ministry of Rural                 Deputy Chairman
          Development
       3.       Secretary of State, Ministry of Planning                        Member
       4.       Under-Secretary of State, Ministry of
          Agriculture Forestry and Fisheries                                    Member
       5.       Under-Secretary of State, Ministry of
          Woman's and Veterans Affairs                                          Member
       6.       Director General, Department General of
          Administration, Ministry of Interior                                 Member
       7.       Deputy Secretary General of CDC, Secretary            Secretary General
          General of CRDB

Article 4.- The chairman of the SEILA Task Force shall have authority to call
            heads/representatives of any institutions and local authorities for the
            meetings in accordance with requirement.

                                         Chapter 3
                                   Roles and Structure

Article 5.- Roles of the SEILA Task Force are:
1. to facilitate the policy discussion on decentralized Planning, Financing and
   Management in the SEILA programme currently being implemented in the provinces
   of Pursat, Battambang, Bantey Meanchey, Siem Reap and Ratanakiri and in the
   other provinces based on available funds;
2. to facilitate the implementation of the SEILA programme among STF Ministry
   Members and the SEILA provinces;
3. to review and approve SEILA Investment Plan and allocate resources;
4. to monitor and evaluate the effectiveness of the decentralized policy implementation
   in the SEILA provinces and prepare reports for submission to the RGC with basic
   recommendations for formulation of a national policy on decentralization throughout
   the country;



                                             40
5. to liaise and cooperate with UNDP and other UN Agencies, Donors and
   collaborating institutions providing technical and financial support to the SEILA
   Programme;
6. to conduct capacity building in planning, financing and management of decentralized
   development.




                                         41
Article 6.- SEILA Task Force shall have a Secretariat located at the CDC.

The roles of the Secretariat are:
1. to facilitate the operations of the STF, preparing annual workplans and organizing
   and preparing the agenda for STF meetings;
2. to maintain and disseminate documents relevant to the implementation of
   decentralization policy;
3. to coordinate tasks and communication among STF Ministry Members, between STF
   and other institutions, and between the national and provincial levels.

Article 7.- The STF Secretariat is managed by a secretary-general of the STF, and
            assisted by deputy secretary-general and some assistants as required.

Article 8.- The staff working at the STF Secretariat shall be any civil servants, who
            would be proposed for secondment from other ministries and institutions.

Article 9.- Establish STF Ministerial Focal Points in the STF ministries to facilitate and
            assist the work of the STF Members.

The roles of the STF Focal Points are:
1. to be a working partner of the STF Secretariat;
2. to prepare ministry workplans relevant to the SEILA programme;
3. to communicate with and support the Ministry's provincial departments in effective
   implementation of the SEILA programme and to resolve any problems falling within
   the Ministry's mandate;
4. to review plans and progress reports submitted by the Ministry's provincial
   departments;
5. to coordinate with other Ministries to ensure consistency and collaboration in the
   SEILA programme;
6. to disseminate results and experiences learned on decentralization within the
   Ministry and prepare recommendations to the STF Secretariat for consideration by
   the STF;
7. to monitor the implementation of the SEILA Programme in the provinces and report
   to the STF member on the status of the activities undertaken by the Ministry's
   provincial departments.



                                           42
Article 10.- The Provincial Rural Development Committee (PRDC), chaired by the
          Provincial Governor, is responsible for the management of the SEILA
          programme at provincial level. In addition to the roles set forth in RGC's
          decision N o 02 SSR dated 11 January 1999 on the establishment of
          Provincial Rural Development Committee, the PRDC's roles are:
1. to review and approve the Provincial Development Plan (PDP);
2. to review and endorse SEILA Investment Plan and submit the investment plan for
   approval by the SEILA Task Force;
3. to approve the allocation of provincial investment funds consistent with the strategy
   of the SEILA programme;
4. to select target areas based on defined criteria and available resources;
5. to promote and support effective cooperation between Government institutions,
   private sector, civil society and local development agencies in planning, financing
   and management of rural development;
6. to coordinate and take action for the implementation of decentralized policy received
   from the SEILA Task Force;
7. to review and agree on the progress reports on the implementation of the Provincial
   SEILA Investment Plan for submission to the SEILA Task Force.

Article 11.- To comply with their roles set forth in Article 10 above, the PRDC shall
            establish an Executive Committee with roles and structure in accordance
            with the STF's guideline to manage the implementation of the annual
            SEILA Investment Plan in the province.

Article 12.- The SEILA Task Force shall use the seal of the Council for the
           Development of Cambodia to stamp all own official letters.

                                        Chapter 4
                                         Funds

Article 13.- Funding sources for the SEILA programme implementation are:

1. Grants and loans from bilateral and multilateral donors;
2. Contributions from the state budget.

Article 14.- The resources mobilized from sources set forth in Article 13, shall be used
             for decentralized rural development named Decentralized Development

                                             43
                    Fund (DDF), and managed by the Ministry of Economy and Finance
                    (MEF). The Ministry of Economy and Finance shall establish the rules and
                    procedure for the operation of the DDF.

    Article 15.- In respect to the management of the funds, the STF Secretariat shall carry
                 out the following tasks:

    1. coordinate with the Ministry of Economy and Finance and Council for Development
       of Cambodia, to mobilize domestic and external resources for the SEILA
       programme;
    2. in collaboration with the Ministry of Economy and Finance, follow up the execution
       of funding agreement signed between the government and donors or lenders and
       submit financial reports according to the agreed schedules;
    3. monitor the transfer of funds from Decentralized Development Fund to the PRDCs
       in provinces for implementing the SEILA programme;
    4. in collaboration with the Ministry of Economy and Finance, conduct auditing on
       PRDC's accounts.

                                                      Chapter 5
                                                    Final Provision

    Article 16.- The sub-decree No 97/ANKR.BK dated 02 December 1997 on
                 establishment of the SEILA Task Force and other regulations set forth
                 against this sub-decree shall be abrogated.

    Article 17.- The Minister in charge of Council of Ministers, Ministers and the Secretary
                 of States of the concerned ministries and institutions set forth in Article 3,
                 and the Governors shall comply with the sub-decree effectively upon the
                 signature.

                                                                            Phnom Penh, 23 August, 1999
                                                                                     Prime Minister
Received place:                                                                    [Signature and seal]
- The cabinet of the King
- Ministry of the Royal Palace                                                              Hun Sen
- Secretariat-general of the Constitution Council
- Secretariat-general of the Senate
- Secretariat-general of the Parliament                               Submitted to the Prime Minister
- Cabinet of the Prime Minister                                             [Signature and seal]
- Council of Ministers
- As set forth in Article 3 "for implementation"                                by Keat Chhon
- Documentation.                                                       Senior Minister, Minister of Economy
                                                                       and Finance and Deputy Chairmen of
                                                          44                         the CDC
ANNEX VII: Sectoral Allocation of Projects

Battambang Province

1996-99
Totals
Type         Number        %             Total Cost %           Local %        Savings %
Building           26               8        76,697        8              13           -3
Irrigation         43              13       121,895       12              17           -2
Road              247              74       756,822       75              15           -4
Water              19               6        48,411        5              12           -9
Non-infra           0               0              0
Total             335                    $1,003,826

1996
Type         Number        Total Cost       Local %
Building               5         8,574           19
Irrigation             5         6,307           20
Road                  22       32,403            19
Water                  0

1997
Type         Number        Total Cost Local %
Building               9       33,074           13
Irrigation             4       19,107           23
Road                  60      236,907           17
Water                  3         9,116          17

1998
Type         Number        Total Cost Local %
Building               8       18,965           10
Irrigation            24       81,537           14
Road                  91      319,110           13
Water                 13       35,267           11

1999
Type         Number      Total Cost Local %        Savings %
Building               4     16,085             10         -3
Irrigation            10     14,944             11         -2
Road                  74    168,402             12         -4
Water                  3       4,029            10         -9
Non-infra              5     46,865              9          0




                                                45
Banteay Meanchey Province

1996-98
Totals
        Type Number    %             Total Cost %        Local %
Building            27             7     86,293     10             12
Irrigation          68            17    125,808     14             18
Road               162            41    370,383     41             24
Water              121            31    257,750     29             16
Non-infra           18             5     59,965      7              8
Total              396                 $900,198

         S
1996
        Type Number       Total Cost Local %
Building              6       15,900           12
Irrigation            4         8,003          12
Road                  7         8,704          24
Water                 4         7,533          10


1997
        Type Number    Total Cost Local %
Building             6     22,884              12
Irrigation          23     48,094              28
Road                49    124,951              29
Water               28     48,932              24


1998
        Type Number    Total Cost Local %
Building            15     47,508              11
Irrigation          41     69,712              13
Road               106    236,728              20
Water               89    201,285              14
Non-infra           18      59965               8




                                               46
Pursat Province

1996-99
Totals
    Type     Number       %         Total Cost   %         Local %
Building          39            8     166,023         20          26
Irrigation        16            3       23,552         3          20
Road             113           24     267,111         32          20
Water            104           22       93,988        11          24
Non-infra        206           43     272,889         33          19
Total            478          100    $823,563        100


1996
    Type     Number    Total Cost    Local %
Building           2        4,824           20
Irrigation        16       23,552           20
Road              24       56,804           20
Water              8       10,119           20
Non-infra         36       30,006           20


1997
    Type     Number    Total Cost    Local %
Building           6        5,805           35
Irrigation         0
Road              12       7,211           21
Water             42      24,454           32
Non-infra          0


1998
    Type     Number Total Cost       Local %
Building          22    74,078              29
Irrigation         0
Road              32    65,039             25
Water             29    33,729             23
Non-infra         84 139073.7              15


1999
    Type     Number    Total Cost    Local %
Building           9       81,315           20
Irrigation         0
Road              45     138,057           13
Water             25      25,686           19
Non-infra         86     103,809           21

                                           47
Siem Reap Province

1996-99
Totals
    Type     Number        %         Total Cost       %         Local %  bid savings
Building          59           12      208,727            29           7         -21
Irrigation        33            7        33,790            5          18         -13
Road             166           35      234,441            33          19         -11
Water            215           45      240,182            33          14          -9
Non-infra           0
Total            473                  $717,141


1996-98
    Type     Number Total Cost        Local %
Building          53  199,227                9
Irrigation        21    21,640              17
Road             126  149,893                8
Water            179  208,703               18
Non-infra          0


1999
    Type     Number     Total Cost    Local %     bid savings
Building           6         9,500           5            -21
Irrigation        12        12,151          19            -13
Road              40        84,548          31            -11
Water             36        31,479          10             -9
Non-infra          0




                                            48
Ratanakiri Province

1997-98
    Type     Number         %         Total Cost     %         Local %
Building          11             15       44,771          16          16
Irrigation          1             1          604           0          31
Road                7             9       44,551          16           5
Water               0
Non-infra         55             74     189,481           68         18
Total             74            100    $279,407          100


1997
    Type     Number      Total Cost    Local %
Building               4     17,000           27
Irrigation             0
Road                   1      2,695              7
Water                  0
Non-infra              7     12,791          27


1998
    Type     Number      Total Cost    Local %
Building               7     27,771           4
Irrigation             1        604          31
Road                   6     41,856           4
Water                  0
Non-infra             48   176,690               9




                                             49
Annex VIII: Correlations between Poverty Indicators




                                                                                                 temporary
                                                                                     % in good




                                                                                                             motorbike
                                                            % boys in
                         % female




                                               displaced




                                                                        % girls in
                                               internally
                         househol


                                    landless




                                                                                                                         cars per
                                               persons
                                    families




                                                                                     housing




                                                                                                 housing
                         headed




                                                            school




                                                                        school




                                                                                                             family


                                                                                                                         family
                                                                                                             s per
                                                                                                 % in
                         ds
                                    %



                                               %
% female headed          1          .094       .015         .045        .015         .015        -.018       .065        -.059
households
% landless families      .094       1          .018         .047        .051         .180        .005        .307        .216


% internally displaced   .015       .018       1            -.013       -.020        -.076       .121        -.045       -.038
persons
% boys in school         .045       .047       -.013        1           .746         .230        -.085       .230        .066


% girls in school        .015       .051       -.020        .746        1            .241        -.079       .236        .053


% in good housing        .015       .180       -.076        .230        .241         1           -.258       .533        .272


% in temporary           -.018      .005       .121         -.085       -.079        -.258       1           -.159       -.092
housing
motorbikes per family    .065       .307       -.045        .230        .236         .533        -.159       1           .436


cars per family          -.059      .216       -.038        .066        .053         .272        -.092       .436        1


rice land per family     .009       -.286      .002         -.021       -.026        -.221       .024        -.276       -.256


**      Correlation is significant at the 0.01 level (2-tailed).
        Correlation is significant at the 0.05 level (2-tailed).




                                                       50

				
DOCUMENT INFO