Project Number: 29257-01
Loan Number: 1753-CAM(SF)
Cambodia: Stung Chinit Irrigation and Rural
Currency Unit – riel (KR)
At Appraisal At Project Completion
23 May 2000 30 June 2008
KR1.00 = $0.000259 $0.000244
$1.00 = KR3,856 KR4,096
ADB – Asian Development Bank
AFD – Agence Française de Développement
FWUC – farmer water user community
GRET – Groupe de Recherche et d’Echanges Technologiques
IRC – Interministerial Resettlement Committee
ISF – irrigation service fee
LBAT – labor-based appropriate technology
MOWRAM – Ministry of Water Resources and Meteorology
MRD – Ministry of Rural Development
O&M – operation and maintenance
PCR – project completion report
PIU – project implementation unit
PMO – project management office
PDWRAM – Provincial Department of Water Resources and Meteorology
SCIC – Stung Chinit irrigation committee
SDR – special drawing right
SMEC SMEC Joint Venture
TA – technical assistance
WUC – water user community
WUG – water user group
In this report, "$" refers to US dollars.
Vice-President C. Lawrence Greenwood, Jr., Operations 2
Director General A. Thapan, Southeast Asia Department (SERD)
Country Director P. Kamayana, Cambodia Resident Mission (CARM), SERD
Team leader P. Long, Project Implementation Officer, CARM, SERD
Team members S. San, Assistant Project Analyst, CARM, SERD
In preparing any country program or strategy, financing any project, or by making any
designation of or reference to a particular territory or geographic area in this document, the
Asian Development Bank does not intend to make any judgments as to the legal or other status
of any territory or area.
BASIC DATA i
I. PROJECT DESCRIPTION 1
II. EVALUATION OF DESIGN AND IMPLEMENTATION 3
A. Relevance of Design and Formulation 3
B. Project Outputs 4
C. Project Costs 5
D. Disbursements 5
E. Project Schedule 6
F. Implementation Arrangements 6
G. Conditions and Covenants 6
H. Consultant Recruitment and Procurement 7
I. Performance of Consultants, Contractors, and Suppliers 7
J. Performance of the Borrower and the Executing Agency 7
K. Performance of the Asian Development Bank 8
III. EVALUATION OF PERFORMANCE 8
A. Relevance 8
B. Effectiveness in Achieving Outcome 11
C. Efficiency in Achieving Outcome and Outputs 11
D. Preliminary Assessment of Sustainability 12
E. Impact 13
IV. OVERALL ASSESSMENT AND RECOMMENDATIONS 13
A. Overall Assessment 13
B. Lessons 14
C. Recommendations 15
1. Project Framework (Original) 17
2. Summary of Project Physical Outputs 25
3. Summary of Vehicle and Equipment Purchases 28
4. Overview of Project Implementation Arrangements 29
5. Status of Compliance with Loan Covenants 33
6. Summary of Resettlement Issues 36
7. Summary of Irrigation Infrastructure Design Chronology: 39
8. Review of Project Economic Analyses 46
9. Sustainability Assessment 52
A. Loan Identification
1. Country Cambodia
2. Loan Number 1753–CAM(SF)
3. Project Title Stung Chinit Irrigation and Rural Infrastructure
4. Borrower Kingdom of Cambodia
5. Executing Agencies Ministry of Water Resources and Meteorology
and Ministry of Rural Development
6. Amount of Loan SDR12,183,000 ($16,000,000)
7. Net Loan Amount SDR10,211,223 ($14,738,557)
8. Project Completion Report Number PCR: CAM 1150
B. Loan Data
– Date Started 2 May 2000
– Date Completed 23 May 2000
2. Loan Negotiations
– Date Started 2 August 2000
– Date Completed 3 August 2000
3. Date of Board Approval 5 September 2000
4. Date of Loan Agreement 12 October 2000
5. Date of Loan Effectiveness
– In Loan Agreement 10 January 2001
– Actual 28 February 2001
– Number of Extensions 1
6. Closing Date
– In Loan Agreement 30 June 2007
– Actual 4 December 2008
– Number of Extensions 1
7. Terms of Loan
– Interest Rate 1.0% during grace period and 1.5%
– Maturity (number of years) 32
– Grace Period (number of years) 8
8. Terms of Relending (if any) Not applicable
Initial Disbursement Final Disbursement Time Interval
23 April 2001 4 December 2008 92 months
Effective Date Original Closing Date Time Interval
28 February 2001 30 June 2007 76 months
b. Amount (SDR million)
Original Revised Amount Amount Undisbursed
Category Allocationa Allocation Available Disbursed Balanceb
01 Civil works, part A 3.60 4.43 4.43 4.06 0.37
02 Maintenance, part A 0.05 0.06 0.06 0.00 0.06
03 Vehicle and equipment, part A 0.25 0.25 0.25 0.21 0.04
04 Consulting service, part A 1.72 2.75 2.75 3.52 (0.78)
05 Training, part A 0.14 0.14 0.14 0.00 0.14
06 Incremental admin, part A 0.64 0.64 0.64 0.62 0.02
07 Civil works, part B 1.69 0.86 0.86 0.70 0.16
08 Maintenance, part B 0.19 0.10 0.10 0.04 0.06
09 Vehicle and equipment, part B 0.18 0.19 0.19 0.20 (0.01)
10 Consulting services, part B 0.36 0.36 0.36 0.25 0.11
11 Training, part B 0.02 0.02 0.02 0.00 0.02
12 Incremental admin, part B 0.40 0.40 0.40 0.29 0.11
13 Interest charge 0.69 0.69 0.69 0.31 0.38
14 Unallocated 2.26 1.30 1.30 0.00 1.30
Total (SDR million) 12.18 12.18 12.18 10.21 1.97
Total ($ million equivalent) 16.00 16.81 17.66 14.74 2.92
The difference between the original amount and the revised total amount is due to the exchange rate variation
between SDR and the US dollar.
The undisbursed loan amount of SDR1.97 million (equivalent $2.92 million) was canceled on 4 December 2008.
10. Local Costs (Financed)
- Amount ($ million): 14.26
- Percent of Local Costs: 55
- Percent of Total Cost: (%) 100
C. Project Data
1. Project Cost ($ million)
Cost Appraisal Estimate Actual
Foreign Exchange Cost 9.60 11.23
Local Currency Cost 14.30 14.36
Total 23.90 25.59
2. Financing Plan ($ million)
Cost Foreign Local Total Foreign Local Total
Asian Development Bank financed 7.70 7.40 15.10 9.77 4.51 14.28
Government of Cambodia financed 4.80 4.80 8.15 8.15
Agence Française de Développement 1.00 1.70 2.70 1.00 1.70 2.70
Beneficiary financed 0.40 0.40 0.00
Subtotal 8.70 14.30 23.00 10.76 14.36 25.13
Interest during Construction Costs
Asian Development Bank financed 0.90 0.90 0.46 0.46
Government of Cambodia financed 0.00 0.00
Beneficiary financeda 0.00 0.00
Subtotal 0.90 0.00 0.90 0.46 0.00 0.46
Total 9.60 14.30 23.90 11.23 14.36 25.59
a. AFD provided parallel grant co-financing of €3.25 million equivalent to about $2.7 million based on exchange
rate in 2001.
Note: Numbers may not sum precisely because of rounding.
3. Cost Breakdown by Project Component ($ million)
Appraisal Estimate Actual
Component Foreign Local Total Foreign Local Total
A. Base Cost
1. Irrigated agriculture development
a. Farmer community organization 0.80 1.60 2.40
b. Irrigation infrastructure 4.60 5.40 10.00
c. System management 0.60 0.40 1.00
01A Civil works 3.23 2.82 6.05
02A Maintenance 0.00 0.00 0.00
03A Vehicle and equipment 0.25 0.02 0.27
04A Consulting service 5.01 0.05 5.06
05A Training 0.00 0.00 0.00
06A Incremental administration 0.26 0.60 0.87
2. Rural infrastructure (part B) 1.50 3.40 4.90
01B Civil works 0.24 0.71 0.95
02B Maintenance 0.03 0.03 0.06
03B Vehicle and equipment 0.26 0.01 0.26
04B Consulting service 0.35 0.00 0.35
05B Training 0.00 0.00 0.01
06B Incremental Administration 0.13 0.27 0.39
Subtotal (A): 7.50 10.80 18.30 9.76 4.51 14.27
1. Physical contingenciesa 0.50 0.90 1.40 0.00
2. Price contingenciesb 0.70 2.60 3.30 0.00
08 Unallocated 0.00
Subtotal (B): 1.20 3.50 4.70 0.00 0.00 0.00
C. Interest Charges 0.90 0.00 0.90 0.46 0.46
07 Interest Charge
Total c 9.60 14.30d 23.90 10.23 4.51 14.74
Physical contingencies are based on 10% of base costs.
Price contingencies are based on average annual escalation of 2.4% for foreign exchange and 6% for local currency
Numbers may not sum precisely because of rounding.
Local currency figures include taxes and duties of about $2.9 million equivalent.
4. Project Schedule
Item Appraisal Estimate Actual
Construction supervision (part A)
Contract date 2001 7 September 2001
Completion 30 June 2007 30 June 2008
Construction supervision (part B)
Contract date 2001 2 May 2002
Completion 30 June 2006
Civil works components
a) Irrigation and drainage infrastructure (part A)
Prequalification and tendering 2001 December 2003
Date of award 6 September 2004
Mobilization October 2004
Start of construction July 2004 15 October 2004
Completion of construction 30 June 2008
b) Rural road maintenance (part B)
Prequalification and tendering
Date of award Mid-2001 31 August 2002
Start of construction 31 August 2002
Completion of construction 30 April 2006
1. 4 February 2004: Approval for minor change of project scope
2. 9 July 2007: Approval of first extension of loan closing date from 30 June 2007 to 30 June 2008
3. 4 December 2008: Approval of cancellation of SDR1,971,777.17 (equivalent to $2,924,914.54)
4. 4 December 2008: Closing of loan accounts.
5. Project Performance Report Ratings
Implementation Period Objectives Progress
From 28 Feb 2001 to 31 Dec 2001 Satisfactory Satisfactory
From 1 Jan 2002 to 31 Dec 2002 Satisfactory Satisfactory
From 1 Jan 2003 to 31 Dec 2003 Satisfactory Partly satisfactory
From 1 Jan 2004 to 31 Dec 2004 Satisfactory Satisfactory
From 1 Jan 2005 to 31 Dec 2005 Satisfactory Satisfactory
From 1 Jan 2006 to 31 Dec 2006 Satisfactory Satisfactory
From 1 Jan 2007 to 28 Feb 2007 Satisfactory Satisfactory
From 1 Jan 2008 to 30 Jun 2008 Satisfactory Satisfactory
D. Data on Asian Development Bank Missions
No. of No. of Specialization
Name of Mission a Date Persons Person-Days of Members b
Fact-finding mission (for loan) 7–25 February 2000 6 d, a, m, e, 2c
Fact-finding mission (for 12–14 March 2000 2 3 d, c
preparation of resettlement plan)
Appraisal mission 2–23 May 2000 4 22 a, d, j, k
Review 1 17–20 October 2000 1 4 a
Review 2 19–23 March 2001 1 5 a
Review 3 2–10 December 2001 1 6 f
Review 4 10–14 December 2001 2 12 a, f
Review 5 2–10 December 2002 1 9 a
Review 6 24–27 March 2004 1 4 a
Review 7 24 February–4 March 2004 1 9 a, g
ADB special mission 8 24 March–4 April 2004 2 12 d, l
Review 9 11–21 May 2004 1 11 a
Review 10 a 24 June 2004 2 3 a, g
Review 11 a 6–17 June 2005 3 5 a, g, c
Review 12 a 24 February–3 March 2006 2 7 a, g
Review 13 a 26–29 September 2006 3 5 a, g, f
Review 14 9–12 May 2007 2 8 a, g
Review 15 9–10 October 2007 1 4 a
Review 16 a 8–9 April 2008 2 4 a, g
Project completion review c 31 August–19 September 2 20 c, g
Fielded concurrently with other missions.
a – project implementation specialist and engineer, b – financial analyst, c – consultant, d – economist,
e – environment specialist, f – assistant project analyst, g – operations and implementation officer, h – transport
specialist, i – senior operations assistant, j – counsel, k – programs officer, l – principal project specialist, m – social
development specialist, n - social development officer.
103o00'E LAO 107o00'E
THAILAND PEOPLE'S DEMOCRATIC
CAMBODIA 14 00'N
RATANAKIRI 14 00'N
MEANCHEY SIEM REAP VIHEAR STUNG TRENG
STUNG CHINIT IRRIGATION AND
RURAL INFRASTRUCTURE PROJECT PAILIN Tonle
(as completed) KAMPONG
THOM KRATIE MONDULKIRI
PURSAT Project Area
Santuk Krao SPEU KANDAL VENG SVAY VIET NAM
11 00'N 11 00'N
Sala Santuk t
km KAMPOT TAKEO
Em PREAH SIHANOUK KEP
o od Mekong R.
North Fl Gulf of Thailand
Stung Tan Flood
Srae Ta Kao Dang Kdar
Veang Tboung Boeng Lvea en t
Khvaek F loo
it Khvaek Open
in Kang Meas
Ch Space Market
Stu South Flood Embankment
Tnaot Chum Khley Chaeung Daeung
Samraong Krava Market
Doun Paen 0 2 4 6 8
Trapeang Chhuok Kilometers
Doun Tom Sralau Toung
Baray Market Chhuk Khsach
Au Soursdey Chan Lhang
KAMPONG CHAM Provincial Town
Rural Market Subproject
Rural Road Subproject
Tradakpong Provincial Boundary
Boundaries are not necessarily authoritative.
I. PROJECT DESCRIPTION
1. The Stung Chinit Irrigation and Rural Infrastructure Project was approved on
5 September 2000 for SDR12,183,000 ($16 million equivalent) and declared effective on
28 February 2001.1 Agence Française de Développement (AFD) approved parallel grant
cofinancing of €3.25 million equivalent to $2.7 million (about 16% of the total project cost2).
2. The rationale for the project derived from a focus on poverty reduction in populous
rural areas, where the mainstay of economic activity was agriculture (constrained by lack of
irrigation, poor roads and market facilities, and limited technology—mainly resulting from the
deterioration of existing irrigation infrastructure). More reliable wet season water flow would
reduce investment risk by avoiding periodic droughts and would (with improved varieties)
allow a second rice crop based on supplementary irrigation. Dry season irrigation would
allow both high-yielding rice varieties and crop diversification. Improved roads would reduce
input supplies and market access costs, and upgraded markets would reduce spoilage and
3. The project goal was to reduce poverty through sustained socioeconomic growth in
an area of Kampong Thom Province that is dominated by rural subsistence or semi-
subsistence farming dependent on one low-yielding, rain-fed rice crop per year. Its purpose
was to increase income and improve the quality of life in the province through provision of
sustainable irrigation, agricultural extension, and rural infrastructure (such as roads and
markets) in and around the project area. The project has two parts: part A—irrigated
agriculture development (components 1–3), and part B—rural infrastructure (component 4).
4. AFD resources were used for component 1 (and some for 3), including for the
consultants Groupe de Recherche et d’Echanges Technologiques (GRET) and Centre
d’Etude et de Développement Agricole Cambodgien. The AFD grant was extended by 1 year
after project closing to support the FWUC and the Chinit Reservoir Irrigation Committee
throughout the next dry season; this support was vital given the construction delays.3
A. Part A
5. Component 1: Farmer community organization and extension services. The
component aimed to ensure that water provision resulted in sustainable increases in
agricultural productivity benefiting local farmers. It involved investments in human resources,
community organization and training, and applied research and extension.
(i) Output 1: Land ownership survey and titling. This output aimed to help
farmers obtain legal title to their lands using maps and ground surveys.
Orthophotomaps4 were to be used for the detailed design of water offtakes
from the secondary canals and for the formation of water user groups.
(ii) Output 2: Water user groups. This output aimed to mobilize, form, and train
water user groups, which were to elect members to farmers water user
communities (FWUCs) for secondary canals. The water user groups and
FWUCs were to be focal points for participation in system design and
operation and maintenance (O&M), and for training in water management.
(iii) Output 3: Agriculture extension services and research. This output aimed
to provide extension services, including farmer field schools and
demonstrations on integrated pest management, improved rice seed and rice
ADB. 2000. Report and Recommendation of the President to the Board of Directors: Proposed Loan to the
Kingdom of Cambodia for the Stung Chinit Irrigation and Rural Infrastructure Project. Manila (Loan 1753-CAM).
AFD financing originally covered component 1 and water quality monitoring and fisheries work in component 3.
It also documented the overall project experience in a post project or capitalization phase.
Orthophotomapping is used to rectify aerial photography by removing distortions caused by terrain relief.
farming techniques, crop diversification, land leveling, animal husbandry, and
post harvest practices. Capacities of provincial departments to provide
extension services were to be strengthened.
6. Component 2: Irrigation infrastructure. The component provided irrigation,
drainage, and associated infrastructure and services for 7,000 hectares (ha). It was to
(i) repair the main diversion weirs on the Stung Chinit and its tributary, the Stung Tang
Krasang; (ii) repair flood bunds, construct new bunds on the eastern side of the main canal,
and provide drainage structures to protect low-lying agricultural areas between the two
rivers; (iii) introduce a fish pass to allow annual fish migrations; (iv) remodel and repair the
main and secondary canals and regulators, along with roughly 60 kilometers (km) of
embankment service roads; (v) provide assistance to farmers constructing tertiary and
quaternary (on-farm) canals; (vi) establish hydrometeorology stations in the Stung Chinit
catchment area; and (vii) provide field offices and equipment.
7. Component 3: Irrigation system management. This was to establish the Stung
Chinit irrigation committee, which was to be responsible for overall management of the
system, and to develop a system for farmer-managed O&M and cost recovery through water
user groups and FWUCs. The capacity of the committee and Ministry of Water Resources
and Meteorology (MOWRAM) staff was to be developed in all technical and administrative
aspects of system management.
B. Part B
8. Component 4: Rural infrastructure. The component aimed to improve about
150 km of rural roads in and around the project area to reduce the cost of transporting inputs
and harvested crops in anticipation of increased volumes of both resulting from greater
agricultural production. Six local markets that lacked permanent roofing, drainage, access to
clean water, truck-loading facilities, and sanitation facilities were also to be upgraded.
9. MOWRAM was the executing agency for part A and the Ministry of Rural
Development (MRD) the executing agency for part B. Interministerial coordination was
provided through a project steering committee chaired by the Ministry of Economy and
Finance, comprising the national project directors from MOWRAM and MRD, and
representatives of various other agencies and ministries.5
10. On 19 December 2003, a major change of project scope was approved, within the
original loan amount and with the goal and purpose of the project unchanged. This reflected
a change in the orientation of the investment to irrigation reconstruction rather than
rehabilitation. The revised design configuration was to provide primary, secondary, tertiary,
and quaternary infrastructure for irrigation, drainage, and rural access in a priority area of
about 2,960 ha (Appendix 1 contains original and revised project frameworks). The
increased irrigation infrastructure cost estimates required (i) the transfer of $1.47 million from
part B to part A ($1.18 million was from loan proceeds and $0.29 million from the borrower)
and (ii) a commitment by the Government of Cambodia for an additional $3 million.
These are the Provincial Department of Agriculture, Forestry, and Fisheries; the Cambodia Agriculture
Research and Development Institute; and the Department of Cadastre and Geography of the Ministry of Land
Management, Urban Planning, and Construction.
II. EVALUATION OF DESIGN AND IMPLEMENTATION
A. Relevance of Design and Formulation
11. The 1995 ADB country operational strategy6 recognized that agricultural and rural
development were key to pro-poor sustainable economic growth in Cambodia. Key areas
identified for improving agricultural productivity include (i) providing and properly maintaining
critical infrastructure for irrigation, drainage, and rural transport; (ii) strengthening agricultural
research and extension; (iii) promoting private sector participation in input supply, product
processing, and marketing; and (iv) providing effective rural financial services. The
government’s First Five-Year Socioeconomic Development Plan7 for 1996–2000 and the
Strategic Plan for 1997–2001 laid a foundation for poverty reduction and economic recovery
focusing on rural development. Key socioeconomic development plan strategies directly
relevant to the project include (i) the achievement of poverty reduction and broad
participation in the development process by focusing on rural development; (ii) substantial
investment to upgrade physical infrastructure, particularly rural roads; (iii) development of the
productive base of the rural economy by raising rice yields for food security and eventually
for export; and (iv) enhanced income opportunities for farm households by diversifying crops.
Increasing liberalization of agriculture markets, ongoing formulation of a water law, and
development of institutional arrangements and procedures to progressively transfer O&M
responsibilities for irrigation systems from the government to farmer water user groups
(based on full participation and ownership principles) provided an appropriate sector policy
context for the project. Notable features of project design are the use of labor-based
appropriate technology in construction methods and the expectation that the government
would view the project experience as a pilot for future irrigation policy reform. The project
design (particularly support to water user groups) benefited from ADB’s own evaluation of
irrigation projects,8 which stressed the importance of technical and managerial training.
12. The high relevance regarding consistency with national development objectives, ADB
strategy, and subsector experience was not matched by the soundness of project design,9
which was characterized by its disjointed and inadequate nature. Investigations under the
first technical assistance (TA) project were interrupted by the deteriorating security
situation.10 It originally suggested that a large-scale irrigation development without a dam
was not viable (which would have been environmentally unacceptable). A second TA project
recommended a reduced scale of irrigation development and added a component for rural
roads and market infrastructure. Partly because of only limited on-the-ground investigations
(and also because of a strong domestic imperative to restore a nationally important11
irrigation scheme) both studies failed to fully appreciate (i) the dilapidated state of existing
irrigation infrastructure, (ii) the sandy nature of many local soils,12 and (iii) the social
complexities of heterogeneous farming operations. There was also something of a technical
disconnection between parts A and B of the project, in that the envisaged generic road and
market improvements did not link very closely to the intended agricultural production
interventions; the nature of the proposed implementation arrangements tended to reinforce
ADB. 1995. Country Operational Strategy Study for the Kingdom of Cambodia: Developing the Capacity for
Reconstruction and Development. Manila.
Royal Government of Cambodia. 1997. First five year Socioeconomic Development Plan 1996-2000. Phnom
Penh. This included a sector development policy statement by the Ministry of Agriculture and Forestry.
ADB. 1995. Sector Synthesis of Post evaluation Findings in the Irrigation and Rural Development Sector.
ADB. 1996. Technical Assistance to the Kingdom of Cambodia for the Stung Chinit Water Resource
Development Project. Manila (TA 2592–CAM); Small-scale technical assistance to the Kingdom of Cambodia
for the Study for Stung Chinit Water Resources Development (TA 3275–CAM) in ADB. 1999. Monthly Report
on Small-Scale Technical Assistance Projects Not Exceeding $150,000 Per Project. Manila.
During appraisal in 1997, Cambodia was still in a post conflict situation after 30 years of civil war.
Originally the scheme was constructed by forced labor under the Khmer Rouge regime.
They are characterized by low organic content, low cationic exchange capacity, and high permeability.
this dislocation. The change of scope enhanced project relevance in some respects, but the
revised canal- and drain-intensive scheme still did not fit well with prevailing socioeconomic
and agro-ecological conditions. Collectively, these design failings almost entirely explain
B. Project Outputs
1. Farmer Community Organization and Extension Services
13.. Land titling was achieved for 2,399 ha, covering 6,138 titles (the project’s physical
outputs are shown in Appendix 2). Farmers received organizational and leadership training,
as well as technical training in accounting and computing skills, to support the FWUC—
made up of farmer representatives covering 25 villages.13 MOWRAM staff provided water
management training courses, extension capacity was developed, and a tertiary
demonstration block was established with a range of rice-based and dry season crop trials.
2. Irrigation Infrastructure
14. The completed irrigation infrastructure covered (i) remodeling and repair of the main
canal from the northern limit of the irrigated area (south of the Stung Tang Krasang) to the
control structure south of the Stung Chinit, including construction of the Stung Chinit spillway
and riparian outlet, and a fish pass to allow annual fish migration; (ii) construction of flood
embankments to the north and south of the Stung Chinit to protect low-lying settled and
cultivated areas; (iii) construction of drainage for protected areas south of the Stung Chinit;
(iv) construction of secondary irrigation and drainage infrastructure, together with associated
embankment service roads and crossing points, in the irrigated area between the Stung
Tang Krasang and the Stung Chinit; and (v) construction of tertiary irrigation and drainage
infrastructure north of the Stung Chinit. Additional works (e.g., control structures, improved
fish passes, box and pipe culverts) were approved under two contracts to improve hydraulic
performance and minimize negative drainage impacts. The project included work to remove
unexploded ordnance (e.g., to remove mines and grenades from the reservoir).
3. Irrigation System Management
15. A system is in place whereby MOWRAM retains responsibility for main canal and
reservoir structures, the FWUC manages secondary and tertiary canals, and farmers are
responsible for quaternary canal management (although using simpler systems of canals
and bunds than originally intended). The original design intentions to have all farmers
supplied individually through a dense network of tertiary and quaternary earth canals resting
on the natural level of the field simply proved too costly in terms of O&M, given the greater
than expected topographic variability and sandy soil. The FWUC also represents farmer
interests at the Chinit Reservoir Irrigation Committee14—established (instead of the expected
Stung Chinit Irrigation Committee) and designed to promote an increased focus on reservoir
and primary and secondary structure (as opposed to a wider river basin remit) management,
and without such an extensive system management authority as originally envisaged.
4. Rural Infrastructure
16. The project improved 84.7 km of road (including seven bridges and 105 culverts) in
10 subprojects to all-weather standard with a 5-meter-wide laterite-paved surface; completed
three rural market subprojects; and constructed an office, workshop, and store at Kampong
Because farmers were found to have multiple plots and undertake many activities other than farming, the only
feasible arrangements for community organization and meetings were on a village-by-village basis.
Established by Municipal Deyka 117, dated 28 November 2007.
Thom. Routine road maintenance was undertaken (2004–2006) for 175 km of roads.
Construction and maintenance generated 450,000 workdays of employment.
17. The major difference between appraisal and completion outputs is due to the
project’s change of scope: the irrigation command area decreased by over half (from about
7,000 ha to 2,960 ha), the length of rural roads was approximately halved, and expected
market upgrades decreased from six to four (of which three were completed).15 The scale
and distribution of project economic benefits were significantly affected by these changes,
mainly through a drop in anticipated cropping, fishery, and vehicle operating cost benefits.
18. Eventual farmer community organization was village based, rather than spatially
arranged, and the absence of quaternary canals reflects the farmers’ choice to have the
scheme operate relatively extensively as opposed to though individual water delivery to each
plot. While an irrigation management system was in place at project closing, its quality (in
terms of comprehensiveness, stability, and sustainability) suffered consequentially from
delays in civil works construction (by reducing the time available to establish a system
involving farmers and government agencies on a fully operating irrigation scheme), and still
awaits formalization via binding agreement between MOWRAM and the FWUC.
C. Project Costs
19. The total cost of the project at appraisal was estimated at $23.8 million. ADB was to
provide a loan of $16 million equivalent (SDR12,183,000). Actual project costs were
$26 million; this difference mainly reflects the increased government commitment of
$3 million following the change of scope. The main shift in cost distribution was to irrigation
infrastructure (under part A) and away from roads and markets construction (under part B).
Foreign exchange costs were higher than expected at appraisal, and local currency costs
lower. SDR appreciation increased the dollar value of the original allocation. The
undisbursed amount of SDR1.97 million ($2.92 million) was canceled in 2008. Appendix 3
summarizes vehicle and equipment purchases by component.
20. Overall fiduciary management was effective and processes in support of project
implementation went relatively smoothly.
21. The first disbursement was in April 2001 (2 months after the loan became effective)
and the last in June 2007. The disbursement schedule was affected by the need to revise
the project scope, and disbursements prior to this were relatively slow. By February 2004
(i.e., after 3 years of loan effectiveness), 26% of the loan had been disbursed; by September
2006, disbursement stood at 61% and by the original closing date of the loan (December
2008) it was just under 90%. Disbursement performance was sometimes affected by slow
submission of commitment letters and withdrawal applications by contractors, and by slow
executing agency review of draft local competitive bidding documents.
22. MOWRAM and MRD operated imprest accounts in accordance with Ministry of
Economy and Finance financial regulations, and ADB’s Loan Disbursement Handbook16 and
guidelines on imprest funds and statement of expenditure procedures.17 Relatively large
initial advances to executing agency imprest accounts and prompt processing of withdrawal
applications thereafter facilitated project operations on the ground. The executing agencies
effectively managed the flow of funds to the project implementation units.
Disagreements between MRD and MOWRAM delayed procurement at one site.
ADB. 2006. Loan Disbursement Handbook. Manila.
Statement of expenditure procedures were allowed for contracts of less than $50,000.
E. Project Schedule
23. The project schedule was severely affected by the need to change the project scope.
The change resulted in an 18-month hiatus in physical activities from late 2002 (once the
dilapidated state of existing infrastructure and the implications of sandy soil were identified)
until early 2004, when the change in scope was specified (on the basis of detailed technical,
social, and economic investigations) and preferred options agreed. The project was
extended by 1 year (to June 2008) to allow for completion of physical works and to support
capacity building of the FWUC and MOWRAM. Savings from the AFD grant in mid-2007
allowed AFD support to continue to mid-2008.
F. Implementation Arrangements
24. The proposed implementation arrangements were broadly appropriate for achieving
the project purpose and delivering outputs, in that they envisaged national coordination
(through the national steering committee), management through ministries (central project
management offices), and implementation at the project sites (through MOWRAM and MRD
project implementation units and involvement of other agencies). Similarly, for the irrigation
scheme itself, the organization of block-based water user groups and their composition into
water user community interface networks with MOWRAM (provincial and national) and other
provincial and district authorities via an irrigation committee to operate the irrigation system
made sense for operations.
25. However, in practice (i) operation and cultural divergences between the "hardware"
and "software" aspects of irrigation system construction and management (which were the
responsibilities of different development partners and groups of consultants) and
coordination platforms established (e.g., the technical coordination committee) to resolve this
were only partly successful; (ii) MOWRAM was reluctant to delegate as much technical
authority to the Provincial Department of Water Resources and Meteorology as was
expected at appraisal, and (iii) implementation coordination between parts A and B was not
as effective as anticipated (the MOWRAM project management office appointed a
coordination officer to address this). Most importantly, the effort and complexities involved in
organizing farmers into a body (the FWUC) to "assume overall scheme management
responsibility" at project closing was underestimated, and the technical and institutional
aspects of farmers’ interaction with state agencies for system management did not evolve as
anticipated. In sum, establishing appropriate functional implementation arrangements around
the revised scope and realities of farmer situations proved a major challenge for the project.
Appendix 4 summarizes the project organization and system management arrangements.
G. Conditions and Covenants
26. Appendix 5 summarizes the extent of compliance with loan covenants. Compliance
with the environmental covenant involved mainly water quality and fisheries monitoring; this
was undertaken satisfactorily. The land survey was completed later than intended.
Resettlement covenants were also complied with, although the scale and complexity of
resettlement issues during implementation rendered their original specification somewhat
simplistic. The establishment of an irrigation system management committee was later than
intended, and the provincial coordination committee did not function entirely as planned with
regard to the frequency of meetings or their ability to meaningfully coordinate technical work
plans. In all these respects, the covenant specification at appraisal and compliance during
implementation were generally acceptable, although the detailed content of several
covenants were simply overtaken by later events. This is particularly true with regard to
covenants related to infrastructure O&M; both the domestic policy references and the
anticipated ADB project-specific practices18 were evolving; as a result covenant compliance
under the project (e.g., regarding formulation of O&M plans, regularizing the use of village
labor) was only partial, although not necessarily inappropriate.
H. Consultant Recruitment and Procurement
27. The project was implemented with assistance from three groups of consultants. The
French nongovernment organization, GRET, provided assistance for component 1 and some
activities under component 3; a joint venture between Lahmeyer International and SMEC
Joint Venture (SMEC) implemented components 2 and 3; and SMEC implemented
component 4. Recruitment was carried out without any difficulties. GRET fielded its
consultants for part A during 2001, a few months later than the Lahmeyer–SMEC
consultants, which meant that the information campaign and farmer feedback on initial
engineering proposals were delayed. Part B consultants were fielded in 2002; overall
contracted inputs were reduced with the change of scope and then later (in 2005) those of
the deputy team leader were extended. Signing of their first contract variation was delayed
somewhat. A total of 114 person-months of international and 418 person-months of national
consultants were used; the final cost was $5.41 million.
28. Contractors, retained through the use of international competitive bidding, undertook
major civil works (i.e., river closure, spillway construction, remodeling main canal) and
control of canal works as envisaged at appraisal. Local competitive bidding was used for
secondary and tertiary irrigation works, and for market and road construction. Skilled and
unskilled labor was employed using force account to supplement these activities. In both
parts A and B, the use of labor-based appropriate technology in large volume proved,
intermittently, more problematic to arrange than envisaged at appraisal.
I. Performance of Consultants, Contractors, and Suppliers
29. Consultants generally performed well, and showed considerable flexibility in
operation and organization as the project scope and implementation schedule underwent
major changes. The engineering quality of completed major civil works is excellent.
30. Quality issues relate to local contractor performance for both parts A and B; the
expected specifications for rural roads subprojects were not always attained and also in
relation to environmental considerations (e.g., the location and condition of borrow pits for
road subbase material). Early ADB missions identified problems in supervision and quality
control, contractor understanding, and contractor management; practice was later improved.
Lack of field coordination between MOWRAM and MRD resulted in detrimental interactions
between rural road and irrigation infrastructure construction work (e.g., in one instance soil
for the road subbase was excavated from places where raised irrigation tertiary irrigation
canals were planned). Within part A, consultants providing the physical infrastructure and
those supporting farmers’ organization had frequent major differences of views; the
preparation of joint terms of reference and overlapping scheduling of individual consultant
inputs only partly addressed the underlying tensions between different perspectives and
J. Performance of the Borrower and the Executing Agency
31. The performance of the borrower and the executing agencies was satisfactory; the
government and the executing agencies (especially MOWRAM) remained committed to the
project's purpose throughout implementation.
ADB. 1995. Report and Recommendation of the President to the Board of Directors: Proposed Loan to the
Kingdom of Cambodia for the Rural Infrastructure Improvement Project. Manila (Loan 1385-CAM[SF]).
32. National institutional capacity was broadly as estimated at appraisal, although
MOWRAM struggled to comply with the demands of ADB safeguards policies and the
conditions of the resettlement plans so as not to delay approval of physical works. The scale
and scope of resettlement issues was not foreseen. MRD commitment to higher project
objectives was somewhat less obvious—the planning and construction of some rural roads
were outside (or only marginally linked to) the project area.19 Central and provincial
MOWRAM and MRD staff were allocated quickly to project operations, and financial and
contracting processes in-country were generally timely, although with some delays in
disbursement and in consultant contracting at the start of the project. The project did not
incorporate any large-scale capacity building.
K. Performance of the Asian Development Bank
33. Overall, ADB’s performance was satisfactory, although with some important caveats.
34. ADB was responsive to borrower and executing agency needs relating to project
initiation and supporting the change of scope in 2003.20 It provided relatively intense
supervision and technical support thereafter, and also with extending the loan period.
Combining the project supervision with that for the Northwest Irrigation Sector Project21and
review of irrigated agriculture issues more generally allowed the policy context of the project
to be dealt with more intensely and increased overall project relevance.
35. Where ADB performance was less than satisfactory is (firstly) with regard to project
monitoring. Specifically, ADB failed to ensure MOWRAM and MRD submitted adequate
project completion reports for parts A and B, and (more generally) staff did not appear to
have any impetus to pursue formal benefit monitoring and evaluation processes. At the time
of loan closing and project completion report preparation, no systematic or rigorous impact
assessment material was on file. Hard and soft copy document storage of the project’s
history at the Cambodia Resident Mission was not very systematic, and throughout the
whole project period consultants’ and executing agencies’ monthly and quarterly reports
generally provide descriptive discussions rather than consolidated up-to-date output
information. ADB performance also appears to have been inconsistent regarding managing
the complexities of resettlement issues. Missions between 2004 and 2006 in particular
seemed to lack consistency of approach and message regarding the preparation of a
stage 2 resettlement plan, or an addendum to the original resettlement plan, and in defining
the detail required in either. ADB had more general tension with MOWRAM (and its own
consultants) over the abandonment of a more "iterative" approach to resettlement planning,
which was understood to have been incorporated within the original resettlement plan.
III. EVALUATION OF PERFORMANCE
36. The project is rated partly relevant.
37. In terms of the project’s fit with the government and ADB sector planning—both
before and after the project—its relevance in 2001 was relatively clear: it fit within broad
government rural development sector objectives and with later, more specific, irrigation and
water subsector policy objectives (e.g., regarding irrigation transfer, implementation of the
Law on Water Resources Management, 2007).
The subsequent MRD proposal for joint physical planning with MOWRAM was welcome, if belated.
Although processing the change in scope required about 6 months after the special administration mission.
ADB. 2003. Report and Recommendation of the President to the Board of Directors: Proposed Loan to the
Kingdom of Cambodia for the Northwest Irrigation Sector Project. Manila.
38. During implementation, this somewhat generalized relevance was enhanced—as
evidenced by the fact that the National Strategic Development Plan 2006–2010 (which is the
overarching framework for reducing poverty through economic growth) is based on the
government's Rectangular Strategy platform, in which a key pillar is enhancing productivity
of the agriculture sector; integral to improving agricultural productivity is the need to manage
and develop Cambodia’s water resources. Thus, the National Water Resources Policy of
2004 (i.e., about midway through project implementation) called for effective, sustainable,
wise, and equitable use of water resources, and the Law on Water Resources Management,
subsequently reaffirmed this policy and adopted more of an integrated water resources
management and development approach. More specifically, the government's present water
resources development priorities include (i) rehabilitating and reconstructing the existing
irrigation and drainage systems (of which Stung Chinit is a major one); (ii) strengthening and
expanding FWUCs by boosting membership and participation of women (i.e., as in
component 1 of the project); (iii) developing and applying measures on flood and drought
mitigation and management (this is also perhaps where Stung Chinit’s usefulness in the
future may increase); (iv) promoting private sector investment in irrigation, drainage, and
other aspects of agricultural water management; and (v) promoting appropriate and effective
river basin management and water allocation systems. The current government's Strategy
for Agriculture and Water22 envisages extensive investment in irrigation management as well
as institution-building to improve scheme management (in partnership with stakeholders).
39. Regarding the evolution of ADB sector objectives as discussed and negotiated with
the government,23 current (2008–2010) country strategy and program outcomes include both
higher irrigation-based production and productivity, and reduced losses from floods and
droughts. Sector outputs to support such outcomes include supporting agriculture sector
reform and improved extension, and direct support for irrigation development.24 Thus the
underlying rationale for irrigation investment—of some type—remains strong. ADB has
sharpened its focus to move away from large direct investments in irrigation schemes,25 not
just in terms of the various Tonle Sap-related initiatives, but also (i) adopting integrated
basin-oriented approaches to irrigation design, and encouragement of water-using farming
communities to sustainably manage small- and medium-sized irrigation schemes; and (ii) a
more general orientation to sectorwide policy reforms and institutional performance
enhancement (to improve service delivery).26 The reason for this latter emphasis derives
from the fact that for much of the project implementation period, the irrigation subsector’s
legal, institutional, and regulatory framework remained in a state of transition, being
governed largely by Circular No. 1 issued by the Prime Minister and Prakas 306 (both of
1999),27 and awaiting promulgation of the Law on Water Resources Management. As a
result, the exact status, rights, and responsibilities of entities in the subsector (and especially
of the FWUC) remained uncertain. The project design anticipated that the project would
serve as a pilot to help refine, revise, and establish a comprehensive legal and policy
environment for water. However, implementation as designed became increasingly
frustrated by the very absence of such a framework;28 as such, its relevance (e.g., being an
investment project rather than a sector development initiative) tended to decrease over time.
Ministry of Agriculture, Forestry and Fisheries and Ministry of Water Resources and Meteorology. 2007.
Strategy for Agriculture and Water 2006–2010. Phnom Penh.
ADB. 2008. Country Operations Business Plan: Cambodia, 2008–2010. Manila.
Including a milestone in the country strategy and program results framework that 35,000 ha of irrigation will be
ADB. 2005. Country Strategy and Program: Cambodia, 2005–2009. Manila.
Expected to be part of the forthcoming Water Resources Management Sector Development Loan (2010).
Circular on the Implementation Policy for Sustainable Irrigation System. This outlined the institutional
arrangements and procedures to progressively transfer O&M responsibilities for irrigation systems from the
government to farmer water user groups (i.e., irrigation management transfer).
ADB missions recommended MOWRAM incorporate in subdecrees:(i) the conditions of membership to a
FWUC; (ii) the operations link between the land title delivery policy and the irrigation service fee recovery
system; (iii) the type and amount of involvement of local authorities; (iv) the system of sanctions to be enforced
40. More directly in terms of project-specific relevance, the choice of investment modality
(i.e., an investment project in a specific geographic area) was probably appropriate at the
time, given this particular scheme’s national importance and the expressed willingness of the
government to treat it as a pilot for subsequent irrigation and water interventions. The logic
for its location near to Phnom Penh (and thus domestic markets) as well of it initially being
import-substituting (Cambodia was still importing rice in 1999) and then later more export-
oriented was sound. The absence of rigorous agriculture sector analysis may be to some
extent understood in the Cambodian post conflict context. Institutional considerations for the
executing agencies were sufficient. The partnership with AFD was a complementary one,
and successfully leveraged both financial and intellectual capital for the project.
41. However, the work used to justify the project investment was not satisfactory, nor was
it based on sound problem-oriented analysis. The underlying technical basis (e.g., for
physical works rehabilitation potential status, soil characteristics) of the original system
design was woefully inadequate. Understanding (at appraisal and at the later change of
scope) of social issues surrounding farming operations, livelihood interests, and farmer
spatial organization, as well as the wider political economy of the water sector was
insufficient. Evidence of the involvement of stakeholders in project design is limited. In
addition, parts A and B of the project lacked economic and operations integration.29
Available decision-support tools for risk and distributional analyses that might have identified
key sources of uncertainty and concern (about both the proposed project design and the
later change of scope) were not employed. An in-depth review of the original design issues,
the attempt to address them by the change of scope, and the causes and consequences of
this situation (including a summary of lessons) is contained in Appendix 7.
42. The 2009 sector assistance program evaluation30 makes similar points, and
concludes that while the case for investment in Cambodian irrigation-based agriculture may
be strong, the flat topography and prevailing soil characteristics tend to limit water
investments in complex lowland agriculture to smaller-scale and technically specific
interventions; this might be true, but was not widely known when the project was designed.
43. One final point concerning the relevance of the project is to note that this may
increase in the future: (i) as the effects of climate change become more apparent, the
capacity to simply store water for agriculture will become more useful, and; (ii) as the
pressure to intensify agriculture increases, the role of sufficient water availability in
supporting both wet season rice yields and dry season diversification may also increase.
44. Regarding the definition of the objectives hierarchy, the project framework
(i) contained means–ends statements (i.e., " … achieved through … ") rather than situation
or condition statements for both the goal and purpose; (ii) stated conditions that should have
been accommodated within project design as assumptions or risks (e.g., regarding
government implementation capacity); and (iii) the lack of specificity in means of verification
perhaps also contributed to the absence of information upon which to assess project impact
in case of unpaid irrigation service fees; (v) the participation of FWUCs in higher bodies for Integrated Water
Resource Management; and (vi) the involvement of FWUCs in agricultural development. To ensure
sustainability of irrigation management transfer, missions recommended MOWRAM review and incorporate in
subdecrees: (i) the different types and magnitude of transfer (partial to total); (ii) the stages and procedures for
a gradual transfer; (iii) the rules and principles for assessing and computing the O&M costs of irrigation
schemes; and (iv) the different contractual modalities for the transfer of the O&M, as well as the financial
commitments necessary to ensure the sustainability of the physical investments.
These issues of relevance arising from project design and formulation would later impact severely on (i) the
implementation schedule; (ii) the extent of outputs delivery; (iii) the effectiveness of outcome achievement, and
(iv) the efficiency of outputs and outcome achievement.
ADB. 2009. Sector Assistance Program Evaluation: Agriculture and Rural Development Sector in Cambodia.
(i.e., because individual studies and surveys were not specified in time-bound terms, they
were not undertaken with any urgency).31
B. Effectiveness in Achieving Outcome
45. The project is rated less effective in achieving its outcome. (A methodological
difficulty makes proper assessment of the extent of achievement of project outcome
difficult,32 given that achievement was defined as provision of both systems and services
(e.g., sustainable irrigation, agriculture extension, rural infrastructure) and simultaneously
consequential and dependent livelihood changes.)33
46. For system and service provision, the project’s effectiveness vis-à-vis appraisal was
necessarily compromised by the reduction in its scale—aggregate delivery of irrigation
command area, rural roads, and markets was about half or less of what was expected at
design. Compared with the change of scope targets, however, full effectiveness was
achieved (irrigation system) or exceeded (rural roads), and a shortfall was experienced only
for rural markets (three markets instead of four)—although delivery of the irrigation physical
works was significantly later than expected. The eventual irrigation system delivery was
somewhat different in quality (regarding its relatively extensive management regimes and
absence of quaternary canals) to what was envisaged at change of scope. Farmer O&M of
the secondary canals and drains is still not assured. Average rice yields are presently below
those expected at project closing, but the trend is going up and may likely be achieved in
another year or two. Significant dry season cropped areas are not yet in evidence, but may
emerge in the future. Local transport costs have been reduced, although the distribution of
benefits (producer and/or consumer surplus) from this is unclear. Side effects of project
implementation may also include positive upstream impacts (reservoir water is extracted for
areas not originally defined as within the project site).
47. In terms of quality of life and income-based measures, the survey does not provide
any evidence that some of farmers’ major livelihood dimensions (financial and physical
assets) have improved; this is broadly confirmed by visual observation and farmer interviews
on the irrigation scheme. At present separating project-induced effects from wider
macroeconomic trends is not possible, especially given that only 1 year has elapsed since
scheme completion and irrigation system management is still evolving. Future expectations
of project effects among local farmers seem to be high, and qualitative survey evidence
suggests the lowest quartile of households may be benefiting most at present.
C. Efficiency in Achieving Outcome and Outputs
48. The project is rated less efficient for achievement of outcome and outputs.
49. No formal evidence is available to undertake a re-estimation of the project’s financial
and economic returns. Detailed discussion of project economic analyses, impacts, and
benefits is in Appendix 8. While the expected economic return was not likely achieved (given
that the scale of outputs was less than envisaged at appraisal, yet project costs increased),
(i) average rice yield productivity achievements are likely to be what was expected, given
that some farmers have already exceeded these expectations (and are producing more than
3 tons/ha); and (ii) the command area shortfall has to some extent been offset by the
doubling of the financial and economic price of rice during project implementation. The net
value of additional rice production may now be around $0.75 million a year. Many local
transport costs have more than halved (and transport services have increased greatly), and
The baseline survey was undertaken in 2005 (i.e., 4 years into implementation).
In the project framework, the terminology for this level of achievement was "purpose."
A more rigorous formulation of outcome would have been in terms of sustainable irrigation-based production
and marketing systems, with increased farmer incomes and poverty reduction defined at the level of impact.
the increase in market trading and market opening periods suggests real efficiency gains
have been obtained. The substantial changes in land values attest to greater overall local
agriculture productivity. The large fisheries element of the revised (at change of scope)
economic internal rate of return calculation is currently much more difficult to estimate; some
observers cite major benefits from reservoir-based fishing, but in an area so inundated with
water for so much of the year fisheries capture from other (and existing) sources is difficult to
separate from reservoir sources.
50. No reason is apparent to consider that project outputs and outcomes could have
been achieved at significantly lower cost. The delivered irrigation infrastructure was
necessary to achieve the kind of water management required to stabilize and increase rice
yields and facilitate dry season cropping. A significant large-scale impact on farmers’
livelihoods in the project area could not have been achieved without introducing a regime of
improved water management. Although the per hectare development cost is relatively high
(at about $3,300 per ha, based on irrigation civil works costs of some $10 million and a
present command area of 3,000 ha), now that this is achieved major efficiency gains are
potentially available by (i) supplementing water management with improved inputs and credit
supply (e.g., to promote activities such as pig raising); and (ii) expanding irrigation command
area (e.g., to the south) at low marginal canal cost, given the reservoir capacity.
51. Civil works construction was of excellent quality and competitively procured.
Consultant costs were perhaps less than usual due to the significant use of nongovernment
D. Preliminary Assessment of Sustainability
52. Project sustainability is rated likely sustainable.
53. Although the other measures of performance are not overwhelmingly positive at
present, features of the project as it now exists make its sustainability quite likely. Among the
factors affecting the likelihood of project sustainability are the following (Appendix 9):
(i) MOWRAM, MRD, and provincial institutions view the project as a significant
success, in terms not just of reconstructing a functioning irrigation scheme to
meet intense local expectations, but also, for example, in terms of lessons on
community organization, participatory system management, and resettlement.
A large amount of domestic political and technical capital has been invested
in the project to date, and the executing agencies are unlikely to allow this to
(ii) The latent potential to relatively cheaply expand irrigated areas around the
project is a prime incentive to maintain the current investment, and currently
planned investments by the government’s own resources, as well as
emerging possibilities for other development partner involvement indicate that
this is fully realized.
(iii) Relatively high commodity (especially rice) prices and the global
intensification of agricultural activities generally (as food demand increases)
will likely keep agriculture operations relatively attractive, and Cambodia is
developing export-oriented food production policies (e.g., the Strategy for
Agriculture and Water 2006–2010), which will facilitate such outcomes.
(iv) Future climate change impacts may result in more droughts and greater
rainfall volume; whatever the actual climate outcomes, wherewithal to better
manage water availability will encourage continuing scheme usage.
(v) Policy affecting issues such as water resources and irrigation management
transfer is still uncertain, but over time should become more settled. To the
extent that the government is able to formalize irrigation-related policy, this
will allow the project system management arrangements (especially regarding
O&M—the single most important determinant of sustainability) to become
clearer and more stable. Sector work undertaken by ADB (e.g., the
preparation of the Water Resources Sector Development loan) will support
this. While the project FWUC situation looks relatively fragile (in terms of its
present and project-limited financial autonomy, and given the short period of
its operations), its general orientation is in line with evolving government
policy, accords with the experience of other FWUCs in Cambodia, and—as
agriculture incomes rise—the scope for increased irrigation service fee
collection (and thus self-financing) may also rise.
54. Socioeconomic impacts. These include crop and income impacts, as well as
increased social mobilization among farmers (in the form of participation in village-based
water institutions through, for example, meetings and elections, and also in leadership
training), as well as greater technical awareness (on water management and agronomic
practices). Positive human resource impacts on the farming community as a whole are
55. Institutional impacts. The impacts on the executing agencies are significant. As
well as dealing with relatively new technical aspects of irrigation development (e.g.,
community mobilization and representation, participatory management, resettlement,
environment), significantly greater coordination and cooperation between ministries (both the
executing agencies and the Ministry of Agriculture, Forestry, and Fisheries and its
departments, e.g., of fisheries and of extension, as well as the Inter-Ministerial Resettlement
Committee) was required than would otherwise have been the case.34
56. Policy and program development impacts. The project experience was viewed as
a pilot from the start by the government/MOWRAM, and although substantiating and
documenting causal relationships is impossible, its experience has likely fed into recent
commitments, for example, to establish a national irrigation maintenance fund,35 and into
wider aspects of participatory irrigation management (e.g., as contained in subdecrees
regulating the FWUC). For both ADB and the government, the project formulation and
change of scope experiences provided the foundation for later Tonle Sap investments.
57. Environmental impacts. Improved water management has reduced the need for
pesticide use. Fish stocks should increase, and the fish pass structure should mean that
migrations are unlikely to be impeded at the spillway. Unexploded ordnance clearance at the
reservoir and construction sites had positive environmental impacts.
IV. OVERALL ASSESSMENT AND RECOMMENDATIONS
A. Overall Assessment
58. Overall, the project is rated36 partly successful. The basis for rating is a weighted
average of the four criteria (relevance, effectiveness, efficiency, and sustainability).
59. The genesis of the project was an attempt to improve rural livelihoods in a very poor
area of Cambodia on the apparent basis of an opportunity to rehabilitate existing irrigation
infrastructure at relatively low cost. Initial preparatory work cast doubt on the investment’s
fundamental viability, but political will for irrigation investment was strong. This was manifest
This is particularly relevant, given intense political competition among Cambodian ministries in the early 2000s.
Reported to be $1 million for 2009 and double this for 2010.
ADB. 2006. Guidelines for Preparing Performance Evaluation Reports for Public Sector Operations. Manila.
in a letter from the Cambodian Prime Minister to the ADB President. The second round of
project design was conducted in an unstable, post conflict environment; topographic, soil,
and social circumstances at the project site were not investigated as fully as would have
been desirable. Data problems were compounded by a lack of rigor in economic analysis,
and the later inclusion of a rural infrastructure component to increase project scale and raise
overall viability made only marginal technical sense. Early implementation experience
revealed that major civil works needed to be reconstructed rather than rehabilitated; the
fundamental choice to be made then was to abandon the project or to redesign it within the
existing loan amount around a (much-reduced) core command area but with features such
that its experience would benefit subsequent irrigation investments. ADB approved a scope
change along these lines; higher objectives were unchanged. There was political pressure to
maintain the investment at this decision point, as elections were being held at about the
same time within Cambodia, and local commitments to irrigation development had been
60. The project’s implementation evidenced significant discontinuity between the
software of community institutional development and irrigation scheme management and
hardware provision, and also between the irrigation and rural infrastructure investments.
Following the change of scope, project implementation was relatively smooth,37 although
farmers had had only very limited experience with operating the scheme at the time of
project completion. Key features of the project as delivered, which differ from expectations,
include village-based (rather than block-based) arrangements for the FWUC and the
absence of quaternary canals on individual plots.38
61. The results of the investment include (i) a major learning experience for executing
agencies, (ii) increased individual and collective technical and management capacity of
farmers, (iii) incremental outputs of rice (and some dry season crops, plus fish), and
(iv) improved incomes and livelihood assets. Perhaps most significant of all is the fact that
the high-quality civil works and the large reservoir capacity mean that the irrigation command
area can now be increased by several times at low marginal cost. The future task for
MOWRAM is to turn what has so far been a pilot project into an irrigation and water
62. The technical and economic implications of investing in irrigation at Stung Chinit were
not fully appreciated by either the Cambodian agencies or ADB. ADB’s TA-based project
preparation and loan processing cycle simply could not handle the complexities and
uncertainties of this situation; either (i) a more intensive design phase was required before
loan approval and ADB management should have insisted on this; and/or (ii) project
implementation should have originally been designed to be more flexible (and local
expectations managed accordingly).
63. The fundamental lesson from the project is that a trade-off exists between eventual
project success based on a robust technical and economic design and the (essentially
political) gains from processing an investment approval primarily to suit a domestically driven
agenda. Proceeding with the project may well have gained ADB credibility in the water
resources sector in Cambodia at the time, but it led to significant implementation and
administration problems in subsequent years. In short, ADB may be criticized for aspects of
The scale and scope of resettlement issues later became problematic.
These proved unpopular with farmers due to on-plot land loss and poor water retention in areas of permeable
soils, and results in a more communal and extensive water management regime than originally intended.
For example, by ensuring adequate O&M on existing works, encouraging development partner or private
sector investment into nearby areas, and facilitating (with the Ministry of Agriculture, Forestry, and Fisheries)
credit and other input supplies.
the design process (at all stages), but also given credit for willingness to engage with
irrigation issues in a difficult political economy environment (as well as for continuing with the
project once its problems had emerged).
1. Project Related
64. Achieving full economic benefits from the project requires that the irrigation
physical infrastructure be maintained and that farmer institutions are able to play as large a
part as possible in scheme management. Future investments in water resources should
specifically include measures to promote adequate O&M of irrigation schemes and formalize
the regulatory and policy environment of irrigation management transfer generally;
continuing such support to Stung Chinit may constitute a possible future subproject
investment. (As MOWRAM is keen to develop irrigable areas adjacent to the project,
technical support to help this objective be achieved in a sustainable and equitable fashion
should also be included). These measures would help ensure that elements of success
under the project are not lost.
65. An appropriate time for a performance evaluation report will be 2013, i.e., 5 years
after civil works completion and when cropping patterns have stabilized.
66. Improving water management is fundamental to enhancing rural livelihoods in
Cambodia, and has a clear fit with ADB and government sector strategies. However, large-
scale irrigation schemes, mainly because of the prevailing flat topography and poor soils, are
technically challenging to undertake. They are similarly demanding because of the intensity
of software support needed for community mobilization at scale. The project design failed to
adequately address these issues, and its implementation and impact suffered
consequentially. But this does not mean that all investments in the sector need suffer the
same fate. Future ADB irrigation interventions could be improved in two major ways: (i) by
focusing on smaller-scale innovations that are capable of wide replication (e.g., the
enhancement of traditional flood and drainage management regimes, and improved dry
season water storage in areas with potential for high-value crops), and (ii) by building on the
benefits of improved physical infrastructure through linked interventions to provide credit and
agriculture technology for new and non-rice activities. Together, this implies more targeted
interventions, based on locally articulated needs.
67. The project demonstrates the necessity to incorporate the broader social
circumstances at all stages of irrigation design and implementation to integrate (and
synergize) hardware and software components. It also highlights the amount of support this
may require. The chances of such integration being achieved will be increased if consultants
from different organizations with widely varying perspectives on project priorities and
objectives are not engaged for the implementation of individual components.
68. Without rigorous formulation of objectives in project frameworks, meaningfully
measuring achievements is difficult, and without specification of clear performance targets
monitoring and evaluation tends to slip. Guidance40 for design and monitoring framework
preparation should fully used. Quarterly and annual reporting (and including the executing
agencies’ project completion reports) by project component should indicate delivery of
outputs to date in relation to targets in the project framework; they should not be qualitative
narratives of recent events.
ADB. 2006. Guidelines for Preparing a Design and Monitoring Framework. Manila.
69. Economic analysis should be wider in its remit than simply cost–benefit analysis; it
should especially be used to investigate alternatives rather than justify one proposed
design.41 The analysis of risk and uncertainty is particularly recommended when the
information design base is limited. The composition of project components should
demonstrate some technical integration and bring economic additionality to the entire
investment, rather than be simply added on to increase scale or apparent viability.
70. Typical project preparatory TA resources need to reflect the project area realities. If
project design is not considered to be robust (at appraisal), then consideration should be
given to building in decision points at key stages of the project such that investment can be
ADB. 2003. Key Areas of Economic Analysis. Manila.
Appendix 1 17
PROJECT FRAMEWORK (Original)
Design Summary Verifiable Indicators Means of Verification Assumptions/Risks
Reduce poverty through • Increased farmer incomes and • Benefit monitoring and • There are no major problems
sustained socioeconomic growth economic activity in Baray and evaluation (BME) surveys, due to natural disasters.
in Kampong Thom Province Santuk districts project completion report (PCR)
• BME surveys, progress reports; • Target beneficiaries will not be
• Increased wet season paddy agronomic observation; local alienated from their land.
production, established dry season district-level agricultural
paddy production, established statistics, PCR
diversified dry season crops • BME surveys, PCR • Implementation capacity of the
• Reduced production, transport, and government is adequate.
Increased incomes and improved • Farmer incomes increased by 10- • Progress reports, review • Price of rice and inputs and other
quality of life in Kampong Thom 40% by 2007 depending on access missions, BME surveys variables remain within ranges
Province through provision of to dry season water. used in economic and financial
sustainable irrigation, agricultural
• Wet season paddy yields increased analyses.
extension, and rural infrastructure
from 1.3 to 3.0 tons per hectare (t/ha) • Stream flow discharges and
(roads and markets)
by 2007; dry season paddy rainfall are sufficient for
production developed over 1,500 ha command area.
by end 2003; diversified dry season
crops developed over at least 500 ha
• Farmer operation and maintenance • Farmers show willingness and
(O&M) of secondary canals and capacity for system
• Average farm-to-market transport • Government policies on O&M
costs are reduced by 40% by 2004. are implemented.
18 Appendix 1
Design Summary Verifiable Indicators Means of Verification Assumptions/Risks
Farmers organized and trained
• Landownership survey and • Public awareness campaign on land • Progress reports, review • Survey and titling can be
titling completed rights completed by June 2001; missions completed before land alienation
landownership survey completed by • BME surveys occurs.
end-2001; undisputed title
applications documented and
• Water user groups (WUGs) submitted by June 2002
and water user communities • About 140 WUGs (each with 50+ • Farmers are enthusiastic for
(WUCs) organized, trained, farmers) and 20 WUCs comprising collective action; leadership of
and granted legal status WUG elected representatives WUGs and WUCs is strong.
established and trained by end-2002;
• Agriculture extension services WUC’s O&M fund established by end
delivered 2003 • Coordination between agriculture
• About 20 commune/village-level and irrigation activities is
extension agents trained by June adequate. Adoption of improved
2002; extension materials and farmer agricultural practices is timely.
• Irrigation infrastructure field schools provided throughout the
developed with input from project period • Full consultations are possible
farmers • 66% of farmers sign off on designs • Progress reports, review within the design period.
for secondary canals and drains, missions
offtakes. • Engineering quality control will
• Irrigation and drainage facilities • Civil works contract requirement be adequate.
to use local labor in project area,
covering 2,000 ha operating by end- • WUGs are able to agree on
2003, remaining 5,000 ha by or WUGs contracted directly
tertiary distribution system and
September 2005 cooperative work arrangements.
• Tertiary canals and drains
constructed by farmers with technical
assistance from the project
Appendix 1 19
Design Summary Verifiable Indicators Means of Verification Assumptions/Risks
Irrigation system management in • Stung Chinit Irrigation Committee • Progress reports; review • Government is willing to share
place and functioning (SCIC) established and functioning missions management responsibility with
by end-2002 • BME surveys farmers.
• WUCs trained to manage secondary • WUGs and WUCs are able to
canal headworks and organize raise sufficient resources,
maintenance by WUGs by mid-2003 including labor, for regular O&M.
• 20 Provincial Department of Water • There is retention of trained
Resources and Meteorology counterpart staff; transfer of
(PDWRAM) and 5 Ministry of Water knowledge from consultants is
Resources and Meteorology adequate.
(MOWRAM) technicians and
engineers trained by end-2002
Rural infrastructure developed • 150 km of rural roads upgraded and • Progress reports, review • Supply of local labor willing to
using labor-based methods six rural markets improved and run missions, PCR/project participate is adequate.
by market user committees performance audit report (PPAR)
• 400,000 person-days of employment • BME survey/PCR/PPAR
Organization and training of $2.8 million • Progress reports, review • Consultants are procured and
farmers, land titling, agriculture • 22 person-months of international missions fielded on time.
extension and 156 of domestic consultants
comprising irrigation institutional
specialists and agronomists
• Local nongovernment organizations
(NGOs) for mobilization and training
20 Appendix 1
Design Summary Inputs Means of Verification Risks/Assumptions
Development of irrigation $12.6 million • Consultants are
infrastructure • Predesign surveys and investigations of • Progress reports, review procured and fielded on
• Organize project structures missions time.
implementation unit (PIU) at • 60 person-months of international and 186 of • Counterpart staff and
PDWRAM domestic consultants, comprising engineers for budget are adequate.
• Recruit consultants hydrological modeling, irrigation, and drainage
• Make surveys, investigations, design, structural design, contracting, and
and detailed design supervision; fisheries specialist, project
• Rehabilitate irrigation and
drainage structures • Office space, support staff, equipment,
Development of system management $1.2 million
institutions and capacity • 17 person-months of international and 16 of
• Progress reports, review • Consultants are
• Establish Stung Chinit irrigation domestic consultants for system water
management and on-farm water management procured and fielded on
committee (SCIC) with full-time
engineering staff; train farmer • SCIC headquarters and meeting halls for
members WUCs • Counterpart staff and
budget are adequate.
• Develop and implement O&M • Training of SCIC members, O&M manuals
plans for WUCs and WUGs • Vehicles
• Test and implement O&M fund
Development of rural infrastructure $6.3 million
• Organize PIU at Provincial • 15 person-months of international and 60 of • Progress reports, review
• Consultants are
Department of Rural domestic consultants comprising a roads missions
procured and fielded on
Development (PDRD) engineer, rural infrastructure engineer, and time.
• Improve 150 km of roads social organizer
• Counterpart staff and
• Re habilitate 6 markets • Office space, support staff, equipment, budget are adequate.
Appendix 1 21
REVISED PROJECT FRAMEWORK (At Change of Scope in 2003)
Design Summary Targets Means of Verification Assumptions & Risks
Reduce poverty through sustained • Increased farmer incomes and economic activity • Benefit monitoring and
socioeconomic growth in Kampong in Santuk district, with some flood irrigation evaluation (BME) surveys,
Thom Province benefits during the wet season in Baray district project performance audit
• Increased wet season rice production, established
dry season rice production, established diversified • Progress reports;
dry season crops agronomic observation;
• Reduced production, transport, and marketing agricultural statistics
Increased incomes and improved • Farmer incomes increased by 10–40% by 2013 • Progress reports, review • Prices of rice and other
quality of life in Kampong Thom depending on access to dry season water missions, BME surveys variables remain within
Province through provision of the ranges used in the
• Wet season rice yields increased from 1.3 to 3.5
sustainable irrigation and agricultural • PCR economic and financial
extension in a priority area of 2,960 tons per hectare (t/ha) by 2013; dry season rice analyses
ha, and rural infrastructure (roads and production developed over 1,350 ha by the end of
markets) 2008; diversified dry season crops developed • Stream flow discharges
over at least 500 hectares (ha) from middle of and rainfall are sufficient
2009 for command area
• Operation and maintenance (O&M) of secondary • Farmers show
canals and drains by farmers willingness and capacity
for system management
• Average farm-to-market transport costs are
reduced by 30% by 2007 • Government policies on
O&M are implemented
1. Farmers Organized and Trained
• Landownership survey and titling • Public awareness campaign on land rights • Progress reports, review • Survey and titling can be
completed completed by June 2003; landownership survey missions completed before land
completed by end-2002; undisputed title alienation occurs
applications documented and submitted by June
2004 • BME surveys
22 Appendix 1
Design Summary Targets Means of Verification Assumptions & Risks
• Water users’ groups (WUGs) and • About 60 WUGs (each with 50+ farmers) and 5 • Farmers are enthusiastic
water users’ communities (WUCs) WUCs comprising WUG-elected representatives for collective action;
organized, trained, and granted established and trained by the end of 2003; leadership of WUGs and
legal status WUC’s O&M fund established by the end of 2004 WUCs is strong
• Agriculture extension services • About 8 commune-village-level extension agents • Coordination between
delivered trained by the end of 2002; extension materials agriculture and irrigation
and farmer field schools provided throughout the activities is adequate;
project period adoption of improved
agricultural practices is
2. Irrigation Infrastructure • 66% of farmers sign off on designs for secondary • Progress reports, review • Full consultations are
Developed with Inputs from canals, drains, and offtakes missions possible within the
Farmers design period
• Irrigation and drainage facilities covering 2,960 ha • Civil works contract
operating by the middle of 2006 requirement to use local • Engineering quality
labor in the project area, control is adequate
• Tertiary canals and drains constructed by farmers or WUGs contracted
with technical assistance from the project directly • WUGs are able to agree
on tertiary distribution
• Up to 600,000 person-days per year (part from system and cooperative
family labor) generated in incremental farm work arrangements
employment, and about 100,000 person-days of
incremental employment in the construction of
tertiary and quaternary infrastructure
3. Irrigation System Management • Stung Chinit Irrigation Committee (SCIC) • Progress reports; review • Government is willing to
Functioning established and functioning by the end of 2005 missions share management
• WUCs trained to manage secondary canal • BME surveys farmers
headworks and organize maintenance by WUGs
• WUGs and WUCs are
by middle of 2006
able to raise sufficient
• 20 technicians and engineers of the Provincial
labor, for regular O&M
Department of Water Resources and Meteorology
and 5 technicians and engineers of the Ministry of • There is retention of
Water Resources and Meteorology trained by trained counterpart staff;
middle of 2006 transfer of knowledge
Appendix 1 23
Design Summary Targets Means of Verification Assumptions & Risks
from consultants is
4. Rural Infrastructure Developed • 80 kilometers (km) of rural roads outside the • Progress reports, review • Supply of local labor
Using Labor-Based Methods priority area upgraded and four rural markets missions willing to participate is
improved and run by market user committees by adequate
the end of 2006 • BME surveys
• 42,000 person-days of employment in road
construction outside the priority area generated
by the end of 2007
1. Organization and Training of $2.8 million
Farmers, Land Titling, • 22 person-months of international and 156 • Progress reports, review • Consultants are engaged
Agriculture Extension person-months of domestic consultants missions and fielded on time
comprising irrigation institutional specialists and
• Local nongovernment organizations for
mobilization and training
2. Development of Irrigation $16 million
• Organize the project • Predesign surveys and investigations of • Progress reports, review • Consultants are engaged
implementation unit (PIU) in the structures missions and fielded on time
Provincial Department of Water • 80 person-months of international and 316
Resources and Meteorology person-months of domestic consultants, • Counterpart staff and
• Recruit consultants comprising engineers for hydrological modeling, budget are adequate
• Make surveys, investigations, and irrigation, and drainage design, structural design,
detailed design contracting, and supervision; fisheries specialist;
• Rehabilitate irrigation and drainage and project accountants
structures • Office space, support staff, equipment, vehicles
3. Development of System $1.2 million
Management Institutions and
• 17 person-months of international and 16 person- • Progress reports, review • Consultants are engaged
• Establish the SCIC with full-time months of domestic consultants for system water missions and fielded on time
engineering staff; train farmer management and on-farm water management
24 Appendix 1
Design Summary Targets Means of Verification Assumptions & Risks
members • SCIC headquarters and meeting halls for WUCs • Counterpart staff and
• Develop and implement O&M • Training of SCIC members, O&M manuals budget are adequate
plans for WUCs and WUGs • Vehicles
• Test and implement O&M fund
4. Development of Rural $4.8 million
• 15 person-months of international and 60 person- • Progress reports, review • Consultants are engaged
• Organize the PIU in the Provincial
months of domestic consultants comprising a roads missions and fielded on time
Department of Rural Development
engineer, rural infrastructure engineer, and social
• Improve 80 kilometers (km) of organizer (these staff will simultaneously be partly • Counterpart staff and
roads engaged on irrigation infrastructure construction budget are adequate
under a portion of part B funds transferred to part A
• Rehabilitate 4 markets
• Office space, support staff, equipment, vehicles
Appendix 2 25
SUMMARY OF PROJECT PHYSICAL OUTPUTS
Table A2.1: Irrigation Infrastructure: Summary Scope of Construction Works
1. FLOOD EMBANKMENT
- Constructed Stung Chinit Flood Embankment in total length 14,448 m
- Constructed Spillway Structure in length 750m 1 place
- Constructed Riparian Structure in length 120m 1 place
- Constructed Fish pass Structure in length 120m 1 place
- Constructed North Headwork Structure 1 place
- Constructed South Headwork Structure 1 place
- Constructed Outlet Structures on flood Embankment 6 places
- Constructed Drainage Canal in length 1,953m
- Constructed South Cross-Drainage Structure 1 place
- Constructed Drainage Canal Headwork 1 place
2. MAIN CANAL
- Constructed of North Main canal earthworks in length 9,742 m
- Constructed Box-culvert crossing Kvek Road 1 place
- Constructed Box-culvert crossing main canal at SD4 1 place
- Constructed Cross-Drainage D2 structure 1 place
- Constructed of Head Regulator Structures 5 places
- Repaired of existing structures on main canal 3 places
3. SECONDARY CANAL
- Constructed of Secondary canals irrigation in total length 20,150 m
- Constructed of Tail Escape Structures end of Secondary canals 5 places
- Constructed of Open Flumes on Secondary canal 49 places
- Constructed of Emergency Spillway on Secondary Canals 5 places
- Constructed of Oxcart Crossing on Secondary Canals 11 places
- Constructed of Box-Culvert on Secondary Canals 3 places
- Constructed of Check Drop Structures on Secondary Canals 9 places
4. SECONDARY DRAIN
- Constructed of Secondary drains in total length 25,461 m
- Constructed of Oxcart Crossing on Secondary Drains 19 places
- Constructed Box-culvert on Secondary Drains 7 places
- Constructed of Drain inlet Protection on Secondary Drain 44 places
5. TERTIARY CANAL
- Constructed of Tertiary Canal in total length 37,926 m
- Constructed of Tail Escape Structures end of Tertiary Canals 39 places
- Constructed of Oxcart Crossing on Tertiary Canals 13 places
- Constructed of PVC pipe Ø 100 Connected from TCs To QCs 1,928 m
6. TERTIARY DRAIN
- Constructed of Tertiary Drain in total length 43,120 m
- Constructed of Check structures on Tertiary Drains 55 places
- Constructed of Oxcart Crossing on Tertiary Drains 2 places
- Constructed of Concrete pipe culvert Ø 300 Connected
- from QDs to TDs 898 m
Source: Ministry of Water Resources and Meteorology (14 September 2009).
TC=Tertiary Canal, QC = Quaternary Canal, Quaternary Drain, TD = Tertiary Drain
26 Appendix 2
Figure A2: Irrigation Scheme as Constructed
STUNG CHINIT PROJECT LAYOUT
0.0 0.5 1.0 2.0
KVAK BOX CULVERT
BOX CULVERT CROSSING NR
REV DATE DESCRIPTION DRWN CHKD APPR
MINISTRY OF WATER RESOURCES
STUNG CHINIT IRRIGATION
RURAL INFRASTRUCTURE PROJECT
Project Layout(North part)
CONSULTANTS NAME SIGNATURE APPROVAL SIGNATURE DATE
APPROVED M.CHEGWIN SCALES : AS SHOWN
DRAWING NO. OOA-LOC 04 B
DATE : 05.08.03
Source: Ministry of Water Resources and Meteorology.
Appendix 2 27
Table A2.2: Rural Road Subprojects
project Work Bridge Culvert Length
No. Road District Commune method (km)
RB01 Krava– Baray Krava Force 2 17 18.0
Veal Spor account
RS01 Kvek– Santuk Kraya Force 1 17 15.5
Dang Kdar account
RB02 Taing Kok– Baray Sou LBAT 1 21 12.3
Au Rumchek Young contract
RB03 Sou Young– Baray Sou LBAT 8 8.6
Tradakpong- Young contract
RS03 Santuk Krao– Santuk Kakaoh LBAT 3 2.7
Sala Santuk contract
RB06 Trapeang Baray Boeng LBAT 2 1.3
RB04 Au Soursdey– Baray Chhuk LBAT 1 4.4
Chan Lhang Khsach contract
RS04 Kang Sav– Santuk Khley Force 4 2.8
RB05 Trapeang Baray Ballang LBAT 14 9.0
RS02 Prasat– Santuk Prasat LBAT 2 19 10.1
Srae Ta Kao contract
Total 7 105 84.7
RB = Road in Baray district, RS = Road in Santuk district, LBAT = Labor-Based Appropriate Technology
Source: Ministry of Rural Development
Table A2.3: Rural Market Subprojects
Subproject No. Market Name District Commune Description
MK01 Baray Baray Baray Standard design for local
market developed under RIIP
MK02 Kvek Santuk Hvek Standard design for local
market developed under RIIP
MK03 Krava Baray Krava Open space market
MK = Market, RIIP = Rural Infrastructure Improvement Project
Source: Ministry of Rural Development.
28 Appendix 2
SUMMARY OF VEHICLE AND EQUIPMENT PURCHASES1
Table A3.1: Components 2 and 3
No. Description Quantity Total Price ($)
1 Civil works (office buildings) 4 152,617.08
2 Vehicles 4 83,500.00
3 Motorcycles 5 18,500.00
4 Heavy equipment 5 72,863.00
5 Information technology 76 85,287.00
− Computers 13 54,470.00
− Sharp copier SF-2540 1 8,694.00
− Sharp plain paper fax FO-885 1 290.00
− HP Design Jet 800PS 1 9,957.00
− CD-R/Rewritable 5 5,050.00
− Printers 5 2,960.00
− Radio communications 6 820.00
− UPSs 13 1,876.00
− Other IT items 31 1,170.00
6 Office furniture 80 7,708.00
− Cupboard, blackboard, cabinet drawers 25 2,590.00
− Map cabinet 2 400.00
− Meeting tables 3 350.00
− Office chairs 21 668.00
− Office desk 29 3,700.00
Source: Ministry of Water Resources and Meteorology.
Table A3.2: Component 4
No. Description Quantity Total Price ($)
1 Vehicles 2 37,000.00
2 Trucks 4 67,738.00
3 Rollers 9 124,950.00
4 Plat compactors 2 3,990.00
5 Water tanks 3 4,200.00
6 Water pump 3 237,878.00
7 Motorcycles 8 9,200.00
8 Desktop computers 4 7,120.00
9 Printers 3 7,900.00
10 Chairs SG 138 16 864.00
11 Chairs OCM-600 40 1,200.00
12 Table AVT2412 2 390.00
13 Chairs LC32c 12 240.00
14 Table DK 2648 and DK 3060 9 950.00
15 Cupboard 304 5 695.00
16 Filings 3 351.00
17 Air conditioners 5 2,662.00
18 Fax machine 1 255.00
19 CD-RW drive 1 290.00
20 Camera 1 450.00
21 Radio communication (ICOM 2001 Japan) 3 786.00
Source: Ministry of Rural Development.
Component 1 was financed by Agence Française de Développement.
Appendix 4 29
OVERVIEW OF PROJECT IMPLEMENTATION ARRANGEMENTS
1. The project design and change of scope envisaged a relatively straightforward
approach to institutional arrangements: national coordination (through the national steering
committee), ministry management (central project management), and implementation at the
site (through project implementation units of the Ministry of Water Resources and
Meteorology [MOWRAM] and Ministry of Rural Development, and involvement of other
2. For the irrigation scheme, the organization of block-based water user groups and
their composition into water user communities interfacing with MOWRAM (provincial and
national) and other provincial and district authorities via an irrigation committee to operate
the irrigation system made sense for operations.
3. The intended implementation arrangements are summarized in Figures A4.1 and
A. Major Institutional Features of Project Implementation
4. In practice, however, during implementation:
(i) an operation and cultural divergence developed between the "hardware" and
"software" aspects of irrigation system construction and management (which
were the respective responsibilities of different development partners and
groups of consultants) and coordination platforms established (e.g., the
technical coordination committee)1 to resolve this were only partly successful
(as was the ability of the provincial coordination committee to manage the
technical and administrative aspects of implementation to the original extent
(ii) MOWRAM was somewhat reluctant to delegate as much technical authority
to the Provincial Department of Water Resources and Meteorology as was
originally expected at appraisal; and
(iii) a shortfall in implementation coordination between parts A and B was
addressed by the appointment of a coordination officer to the MOWRAM
project management office.
5. Most importantly, however, (i) the effort and complexities involved in organizing
farmers into a body (the farmer water user community [FWUC]) to "assume overall scheme
management responsibility" at project closing was underestimated; and (ii) the technical and
institutional aspects of farmers’ interaction with state agencies for system management did
not evolve as anticipated.
6. In sum, establishing appropriate functional implementation arrangements around the
revised scope and realities of farmer situations and the national institutional landscape
proved a major challenge for the project.
This was essentially an unenvisaged attempt to integrate and reconcile for MOWRAM absorption the various
streams of technical information and opinion generated by implementing consultants under components 1–3
and the tertiary units under the project implementation unit.
30 Appendix 4
Figure A4.1: Project Organization and Coordination
MOWRAM MEF Interministry MAFF MRD
Undersecretary (Chair) CDC MLMUC Resettlement Undersecretary Undersecretary of
of State Committee of State State
Executing Project Project
Agencies Management Office Management Office
accounts) ADF-and ADB-
Office of Stung Chinit Stung Chinit PDAFF PDRD
Project Project Provincial Coordination
the Irrigation Resettlement. Office of Project
Implementation Implementation Department Committee
Governor Committee Committee Extension Implementa
Represent Represent. tion
Components/ Irrigation System Farmer Benefit Monitoring Rural
Activities Infrastructure Management Organization / Ext. and Evaluation Infrastructure
---------------------------Part A----------------------------------- Part B
ADB = Asian Development Bank, AFD = Agence Française de Développment, CDC = Council for the Development, MAFF = Ministry of Agriculture, Forestry and Fisheries,
MEF = Ministry of Economy and Finance, MLMUC = Ministry of Land Management, Urban Planning and Construction, MOWRAM = Ministry of Water Resources and
Meteorology, MRD = Ministry of Rural Development, PDAFF = Provincial Department of Agriculture, Forestry and Fisheries, PDRD = Provincial Department of Rural
Development, PDWRAM = Provincial Department of Water Resources and Meteorology.
Source: Ministry of Water Resources and Meteorology.
Appendix 4 31
Figure 4.2: System Management Arrangement
Water Water Water Water Water • Operation of canal offtakes
50-hectare User User User User User • Water management and dispute
blocks Group Group Group Group Group resolution for 50 ha blocks
• Maintenance of tertiary canals and
drainage, and field-to-field irrigation
Elected Elected • Collection of fees (in cash or in kind)
• Operation of secondary canal head
Secondary Water User Water User works, canal dispute resolution
canals Communities Communities • Organization and management of
maintenance of secondary canals
• Management of water user fees
• Representation on the Stung Chinit
Stung Chinit • Overall scheme management
Overall Irrigation Committee • Through MOWRAM technical staff,
scheme Chaired by WUC Rep. operation of main canal weirs and
management Majority membership control structures
from WUCs • Operation and maintenance
Technical training • Set water user fees
Implementation MOWRAM and PDWRAM District and provincial • Technical support
consultants technical staff and project authorities
MOWRAM = Ministry of Water Resources and Meteorology, PDWRAM = Provincial Department of Water Resources and Meteorology, WUC = water user community, WUG =
water user group.
Source: Ministry of Water Resources and Meteorology.
32 Appendix 4
7. Three factors made the organization of farmers into a body to assume partial system
management responsibilities more difficult than originally anticipated: (i) the multiple
agricultural and nonagricultural activities of many farmers, which meant that many
individuals had other demands on their time such that willingness and ability to participate in
village and farmer meetings (and to be involved in any eventual farmer organization) was
reduced; (ii) because hydraulic scheme boundaries, administrative boundaries, and farmers’
plot distributions do not necessarily coincide, any single irrigation block (bounded by
secondary and tertiary canals) does not exhibit enough social cohesion upon which to build
sustainable irrigation management capacity, and (iii) because many farmers have plots
scattered across several blocks means that an individual would be required to attend several
8. These problems could have been overcome to some extent by designing tertiary
canals to follow village boundaries, but in general the original approach to block-based
farmer organization was essentially inappropriate. The eventual FWUC organization is
village-based for its composition and representation.
9. The FWUC is now a "professionalized stakeholder," comprising a director, irrigation
operator leader and agents (16), plus a scheme maintenance leader and agents (8, plus a
village facilitator, accountant, and irrigation service fee collection team—usually village
leaders. Farmers elect their village representatives and the five members of the FWUC
committee. FWUC staff have received training in leadership facilitation and organization (via
Groupe de Recherche et d’Echanges Technologiques (GRET) and Centre d'Etude et de
Développement Agricole Cambodgien CEDAC), as well as accounting and computing, in
addition to water management (from PDWRAM staff). The effort required for creating and
strengthening the FWUC in fact came to occupy most of the software side of project
implementation (especially as the arrangements for wider system management remained
constrained by the wider legal and policy environment).
10. The evolution of the project FWUC was a tortuous process throughout project
implementation, as regards the definition of its functions, responsibilities, composition, and
organization. FWUC status was legally recognized in 2006.
11. As regards the second point, the irrigation scheme management system integrated
farmer and state responsibilities on agreed technical and administrative principles, the
original intention was that a system would be established in line with Circular No. 1 of 1999
on the implementation policy for sustainable irrigation schemes. While this provided general
principles for irrigation management transfer, it did not fully facilitate operating practice.
Increasing participatory aspects of irrigation management transfer did not prove easy for the
project given the fairly conservative interpretations of the circular by the Participatory
Irrigation Management Department in MOWRAM.
12. At project closing in mid-2008, a fully defined shared management system for the
irrigation scheme was still not in place. The letter of agreement was signed in October 2009
(under the auspices of the Chinit Reservoir Irrigation Committee; it will be recognized under
a MOWRAM subdecree) that defines (i) the respective responsibilities of the FWUC and
MOWRAM regarding management of operation and maintenance across the range of
scheme infrastructures; (ii) ownership of property; (iii) funding estimates of and financial
responsibilities for O&M; and (iv) monitoring and dispute resolution procedures. The letter of
agreement is for a transitional period of 3 years during which the FWUC is expected to
increase irrigation service fees and/or raise funds from external sources (i.e., MOWRAM,
Provincial Department of Water Resources and Meteorology, development partners) to meet
its operation and maintenance obligations (A more complete discussion of the implications of
this for sustainability is contained in Appendix 9).
Appendix 5 33
STATUS OF COMPLIANCE WITH LOAN COVENANTS
Project-Specific Date Date Delays Status of Rating
Covenants Due Complied (mo.) Compliance (HS,S,PS,U)
Environmental Not Yet Due
1. Loan Agreement (LA), Schedule 6, Para 8. The Borrower will ensure that all environmental mitigation measures set out in the
summary initial environmental examination report are incorporated in the project designs and are implemented.
Complied With S
1. LA, Sch. 6, Para 9(a). Resettlement. The Borrower will ensure that all people affected by the Project are compensated in a
timely manner in accordance with ADB’s Handbook on Resettlement and compensation measures set out in the resettlement plan.
Upon completion of the detailed design of the irrigation infrastructure works, but prior to the start of any construction of such
works, the Interministerial Resettlement Committee (IRC) will submit to ADB for approval an updated resettlement plan.
Complied With S
2. LA, Sch. 6, Para (b). The IRC will be responsible for coordinating the implementation of the Resettlement Plan at the national
level. The Stung Chinit Resettlement Subcommittee (SCRS) will be responsible for implementing the resettlement plan under
supervision of the IRC.
Complied With S
3. LA, Sch. 6, Para 9(c). The IRC through MOWRAM will keep ADB informed of the progress of implementation of the
resettlement plan through quarterly progress reports and through a report to be submitted immediately following completion of
implementation of the Resettlement Plan and six months thereafter.
Complied With S
4. LA, Sched. 6, Para 10: Land Titling. The Department of Cadastre and Geography under the Ministry of Land Management,
Urban Planning, and Construction will be responsible for implementing part A (i) of the Project. The Borrower will ensure that the
Department of Cadastre and Geography will (i) by January 2002, have completed a survey of ownership of all land in the project
area, and prepared and publicly posted cadastral index maps in all villages in the project area; and (ii) by June 2002, have
completed the technical inspection of all undisputed parcels for the adjudication record.
Complied With Late S
5. LA, Sch. 6, Para 11. The project implementation unit (PIU) for part A will design the secondary canals and offtakes and tertiary
block boundaries in consultation with the farmers affected by such design. The overall design of the irrigation structure will be
completed only after at least two thirds of the farmers along each secondary canal have approved the designs for the secondary
canals, offtakes, and tertiary block boundaries.
Complied With HS
6. LA, Sch. 6, Para 12. The Borrower will ensure that public markets under part B are constructed only on existing market sites or
public land that is not occupied or used otherwise by private persons. In selecting roads and markets for upgrading under part B,
MRD will give strong preference to the views of target beneficiaries through surveys and meetings at the village, commune, and
Complied With S
7. LA, Sch. 6, Para 17. Before implementation of any market upgrading activities, a market committee for the market concerned
will be established and market user fees to be charged shall be determined. The composition of the market committee, its terms
of office, and the market user fees will be subject to approval by the Provincial Rural Development Committee. The market
committees will be responsible for O&M of the market facilities. The market user fees will be sufficient to cover regular O&M and a
fund for further maintenance and improvement of the market.
Partly Complied With S
1. LA, Sch. 6, Para 15. MOWRAM and MRD will ensure that annual maintenance plans are prepared for rural roads and the
main irrigation canal and control structures on the basis of physical measurements of needs and agreed-upon maintenance
standards. The Borrower will allocate sufficient budget to cover O&M costs.
Complied With S
34 Appendix 5
2. LA, Sch. 6, Para 16. For the irrigation system, the Borrower will ensure that O&M is carried out in accordance with the
principles of its Implementation Policy for Sustainable Irrigation Systems. For rural roads, the Borrower will apply the system
and financing arrangements for routine and periodic maintenance of rural roads to be developed under the ADB-financed
Rural Infrastructure Improvement Project.
Partly Complied With S
3. LA, Art., IV, Sect. 4.06(b). The Borrower will maintain, or cause to be maintained, separate accounts for the Project and
have them and related financial statements audited annually. The Borrower will furnish to ADB not later than 12 months after
the end of each related fiscal year (FY), certified copies of such audited accounts and financial statements and the report of
the auditors relating thereto.
Complied With HS
1. Established, Staffed, and Operating PMU/PIU part A: LA, Sch. 6, para. 1. National Level. MOWRAM will ensure that
throughout project implementation the project management office (PMO) for part A will be headed by a national project
director at the level of undersecretary of state and include a full-time manager at the level of director.
Complied With HS
2. Fielding of Consultants: L.A., Art. IV, Sect. 4.03(a): The Borrower will cause competent and qualified consultants and
contractors, acceptable to ADB, to be employed.
Complied With HS
3. LA, Sch.6, Para 2. Provincial Level. MOWRAM will ensure that throughout project implementation the PIU for part A will
be headed by the provincial director of water resources and meteorology as project manager and include a deputy project
manager and an adequate number of suitably qualified staff on a full-time basis.
Complied With HS
4. LA, Sch. 6, Para 3. Prior to 30 June 2002, MOWRAM will establish the Stung Chinit Irrigation Committee (SCIC), which
will be responsible for managing the Stung Chinit irrigation system. By 30 June 2003, MOWRAM will have completed
construction of a headquarters building for the SCIC.
Complied With Late PS
5. Part B. LA, Sch. 6, Para 4. National Level. MRD will ensure that the PMO established for the Project will continue to
function throughout project implementation as the PMO for part B, under the direction of a national project director at the
level of undersecretary and with a full-time national project manager and including additional staff as needed.
Complied With S
6. LA, Sch. 6, Para 5. Provincial Level. MRD will ensure that throughout project implementation the PIU for part B will be
headed by the provincial director of the Department of Rural Development as project manager and will include a full-time
deputy project manager and an adequate number of suitably qualified staff. MRD will provide three road construction
teams comprising a technical adviser, provincial engineer, site engineer, and three supervisors to support the PIU.
Complied With S
7. LA, Sch. 6, Para. 6. National Level. Within 30 days of the effective date, the Borrower will have established a project
steering committee (PSC) and appointed a representative from the Ministry of Economy and Finance (MEF) as
chairperson, and National Project Directors from MOWRAM and MRD and representatives of the Ministry of Agriculture,
Forestry, and Fisheries (MAFF); the Ministry of Land Management, Urban Planning, and Construction; the Cambodia
Development Council; and the chairman of the IRC as members. The PSC will meet at least semiannually, with the project
managers present, and will be responsible for overall coordination of project implementation among the project executing
agencies and dispute resolution.
Complied With S
8. LA, Sch. 6, Para. 7. Provincial Level. Within 30 days of the effective date, the Borrower will establish a Project
Coordination Committee. The committee will meet at least quarterly and will be responsible for coordinating
implementation activities and adjusting work programs as needed.
Partly Complied With PS
9. LA, Art. IV, Sect. 4.07(b): The Borrower will furnish, or cause to be furnished, to ADB quarterly reports on the carrying
out of the Project and on the operation and management of project facilities.
Appendix 5 35
Complied With HS
10. LA, Art. IV, Sect. 4.07(c). Promptly after physical completion of the Project, but not later than 3 months thereafter, the
Borrower will prepare and furnish to ADB a report on the execution and initial operation of the Project.
Partly Complied With PS
11. LA, Sch. 6, Para. 13. Except as the Borrower and ADB may otherwise agree, the following criteria will apply in the
selection of subprojects under part B: (i) the estimated subproject cost will be less than the equivalent of $450,000 (in
2000 prices), and (ii) the expected implementation period for each subproject will not be more than 2 years.
Complied With HS
12. LA, Sch. 6, Para. 14. MRD will submit all subprojects proposals to ADB for approval prior to their implementation, until
the Borrower and ADB have agreed that MRD will be responsible for final approval of subprojects. If, exceptionally, a
subproject is estimated to cost more than the equivalent of $450,000 (in 2000 prices), this subproject will continue to be
subject to approval by ADB prior to implementation.
Complied With S
Overall Rating S
(–) = no rating available. FY = fiscal year, HS = highly satisfactory, PS = partly satisfactory, S = satisfactory, U = unsatisfactory.
36 Appendix 6
SUMMARY OF RESETTLEMENT ISSUES
1. Three stand-alone but closely linked resettlement plans were prepared for the project;
the first (resettlement plan phase 1) for the irrigation and drainage canal infrastructure; the
second (resettlement plan phase 2) for the reservoir area; and the third (resettlement plan
phase 3) for the Ochork Tributary area.
2. The plans identify a total of 3,685 affected households: 1,778 under resettlement plan
phase 1; 1,103 under resettlement plan phase 2; and 804 under resettlement plan phase 3.
During the course of the implementation of each plan, additional households came forward with
claims for damage to crops and farmland due to flooding during the rainy seasons in 2006 and
2007, which were claimed to have been caused by the construction of various components of
the irrigation system.
3. Overall, the need to change the project scope led to the creation of an increasingly
complex and contentious resettlement situation that ADB and local institutions struggled to fully
deal with throughout project implementation.
A. Project Design Resettlement Expectations
4. At project preparation, preliminary canal design work indicated that about 110 hectares
(ha) of government-owned land used for recession rice would be inundated once the weirs were
repaired, and the size of the inundation zone would be expected to decline steadily from the end
of the rainy season through the dry season, allowing some households to continue recession
rice agriculture, but delayed by 1–2 months. The primary users of the land came from a nearby
village of 73 households. Although no settlements were to be flooded and no physical relocation
of households was required, people then currently using land that would become permanently
inundated would receive compensation (in the form of alternate land for agriculture).
5. The impacts from canal rehabilitation were originally expected to be minor because the
rights-of-way were already established, except for 22 households located on the eastern bank of
the main canal, near the Stung Chinit weir (these households were to be relocated to an area
near the Stung Chinit river to preserve their primary economic activity of bamboo collection
upstream). Rural roads were to use the existing alignments, and markets were to be built on
existing market land or unoccupied public land donated by the district or commune authorities;
they were not expected to have adverse resettlement impacts.
B. Change of Scope Resettlement Implications
6. At the change of scope, based on the preferred design option, the total number of
households affected (in the priority area) was provisionally calculated at 1,778. This significant
change in number was explained by (i) the need to reconstruct, not repair, existing irrigation and
drainage infrastructure; (ii) confirmation by the Ministry of Water Resources and Meteorology
(MOWRAM) and the Interministerial Resettlement Committee that the Land Law of 30 August
2001 established a legal requirement, not recognized at loan appraisal, to compensate for land
lost to secondary canals and drains; (iii) more accurate topographical studies and
orthophotomapping than were available at loan appraisal; and (iv) inclusion of affected peoples
in the impact area of flood embankments and drains required by the Stung Chinit reservoir.
Appendix 6 37
7. Of these affected people, about three-quarters were to suffer only marginal land loss or
some loss of seasonal land use that could be overcome by changed land and water use. The
project was not expected to cause major displacement of people or houses. A total of 39 houses
and 28 trading stalls would be displaced on the main canal, and 16 houses would be displaced
by the reservoir, all of which were to be relocated close by.
8. By September 2007, an additional 373 households (under resettlement plan phase 1)
claimed that their crops and farmland were affected by the project; by October of the following
year, 352 of these additional affected households had been compensated (21 others were
confirmed to have no valid claim for losses). In addition, 33 landless affected households were
living along the canal embankments and needed to be relocated; a relocation site was provided
in Kvek village. (Subsequently, 1 of the 33 relocated households was found to be still living at
the site, the others had left and apparently sold their allocated home plots to village residents.)
C. Experience after Implementation of the Change of Scope
9. During implementation, all affected persons were compensated promptly for loss, with
the exception of 383 who were severely affected by unexpected flooding during the 2006
season and waited a year for compensation. Some residents of Snao and Kley villages
experienced flooding due to poor drainage and waited a long time for compensation of poultry
and garden loss. More significantly, the construction of the reservoir necessitated revision (in
the form of an addendum) to the approved resettlement plan, and this included more than 1,100
affected persons, of which 777 were severely affected through loss of 10% or more of their
productive assets. (By September 2007, an additional 316 affected households under
resettlement plan phase 2 were also identified). Compensation payments were completed by
10. In Kampong Sdach village, about 13 households were seasonally inundated and
arranging compensation proved problematic as some households did not wish to relocate away
from the water’s edge. A relocation site was prepared located about 1 km from the existing
location of the 13 affected households, and the affected households accepted compensation at
replacement cost for their houses. By the end of 2008, however, only three affected households
had moved to the relocation site; the 10 others wanted to keep both their present location and
their replacement plots in the relocation site. The 10 households are reportedly engaged in the
trade of illegally cut logs and use the creek to transport the logs. In addition, another 300 or so
households were unexpectedly affected by the 2007 flooding. Further problems arose with the
flooding of the Ochork tributary, and resulted in about 650 households losing permanent use of
some agriculture land.
11. As well as these changes due to the direct, if unexpected, project impacts, issues
surrounded the compensation for different types of canals (land loss to secondary canals and
drains was expected to be compensated, but land loss to tertiary and quaternary canals would
not be—this seemed unfair to some farmers). Issues of compensation in the form of land were
complicated by (i) increasing purchase of land by local (and not so local) speculators; (ii) by the
inherent limitation of suitable paddy land; and (iii) by farmers’ fundamental distrust of the
administration and dispute resolution system in the event of conflict. These factors collectively
tended to make farmers prefer cash compensation where possible.
12. As the whole resettlement situation became increasingly complex during 2005–2006, a
change of staff within ADB and confusion between ADB and MOWRAM (who had their own in-
house international consultant) developed about what would be required in terms of content and
38 Appendix 6
detail for revisions to the resettlement plan, and also a divergence of opinion about what the
actual project impacts had been at Kampong Svay. MOWRAM regards this period as one of
discontinuity and poor institutional memory on the part of ADB, and especially unfortunate in
resulting in the discarding of detailed information compiled by MOWRAM for a draft addendum
(which ADB had initially requested) in favor of preparation of a phase 2 resettlement plan.
13. While the project (at ADB headquarters, at the Cambodia Resident Mission, within the
MOWRAM resettlement unit) has generated an enormous amount of material covering
resettlement planning, personal memories of institutional and staff may conflict and be
incomplete. Some points emerging from the project experience include the following:
(i) The original design of the project without significant social development
involvement led to a fundamental lack of appreciation of the likely complexities of
resettlement planning vis-à-vis land availability, the need to fend off land
speculation, compensation for different canal types, and farmers’ preferences for
cash (against ADB’s standard policy of preference for alternative land).
(ii) The 2003 change of scope necessarily expanded the scale of resettlement; about
$3.26 million was eventually spent on resettlement.
(iii) The early expectations of MOWRAM were that an essentially evolving or iterative
approach would be used for resettlement planning, rather than a need to prepare
new resettlement plans as ADB later required—this caused significant friction
between MOWRAM and several resettlement missions conducted by ADB. What
resulted from this was a perception within MOWRAM (and its own international
consultants) that ADB resettlement supervision is "spotty" and infrequent, and is
poorly resourced, overly formulaic, and time-driven, as well as hierarchical and
(iv) Both continuity of personnel (desirable) and clarity of process (essential) are
required for good project administration, and in these conditions it is arguable
whether either were in place on the part of ADB.
(v) Notwithstanding these issues, the implementation experience of the project did
articulate a major formal resettlement procedure in Cambodia.
Appendix 7 39
SUMMARY OF IRRIGATION INFRASTRUCTURE DESIGN CHRONOLOGY:
CAUSES, CONSEQUENCES, AND IMPLICATIONS
1. The development of the irrigation infrastructure for the Stung Chinit Irrigation and Rural
Infrastructure Project is best divided in three epochs. The first is from 1996 to 2000, a period of
early technical assessments and high expectations, where the "unknowns" had not been
identified. The second is from 2001 to 2003, a period of project mobilization and discovery,
where the unknowns were discovered and articulated, and a change of scope was required to
address the realities of the operating environment. The third and final is from 2003 to the
present day, when an options strategy was formulated, a course of action decided upon, and an
irrigation system constructed which is operating today (albeit at a reduced command area) with
the potential for future expansion.1
A. 1996–2000: The Project Formulation
2. Initially a project preparatory technical assistance (TA) study was prepared for Stung
Chinit Water Resources Development, and approved in June 1996.2 The project as requested
by the government was for a new upstream reservoir on the Stung Chinit River to provide wet
and dry season irrigation supply for 12,000 hectares (ha).
3. At the time, the Asian Development Bank (ADB) was concerned about environmental
impacts (especially of a dam in neighboring uplands) and local capacity to implement the
upstream reservoir project. ADB proposed a less ambitious option, involving rehabilitation of an
existing irrigation system using weirs across the Stung Chinit and Stung Tang Krasang to
manage run-of-river flows to irrigate 12,000 ha in the wet season and 2,000 ha in the dry
season.3 The quality of the feasibility study report prepared under the TA was rated poor, partly
because the TA was interrupted in 1997 due to a deterioration in security.4 The deterioration in
security made it impossible to inspect the site and, as a consequence, the condition of the
irrigation structures was unable to be determined. 5
4. A follow-up small-scale TA was initiated in 1999 at the government's request to reassess
the proposed project.6 The TA recommended a downstream weir with a reduced irrigation
command area of supplemental wet season irrigation for 7,000 ha and dry season irrigation for
2,000 ha. Engineers determined this area was more appropriately matched to potential storage
5. The decision to scale the project down—from 12,000 ha with an upstream reservoir to
7,000 ha with weir and spillway—was an improvement. However, the findings from this TA
This report focuses primarily on the irrigation infrastructure impacts on decision making. Clearly other factors are
influencing technical decisions to proceed, most importantly the political economy of the time.
ADB. 1996. Technical Assistance to the Kingdom of Cambodia for the Stung Chinit Water Resource Development
Project. Manila (TA 2592-CAM). The original study aimed to assess a project proposal prepared by the
Government of Japan as early as the 1960s.
The existing irrigation system on the Stung Chinit was established in the Khmer Rouge era with the weir structures
still standing. Between 1978 and 1989 the system was operating; however, in 1989 a large flood breached the
earth dike and the system fell into disrepair.
The deterioration in security made inspection of the site impossible and, as a consequence, the condition of the
irrigation structures could not be assessed.
Stung Chinit Irrigation and Rural Infrastructure Project, Change in Project Scope Report, 2003.
Small-scale technical assistance to the Kingdom of Cambodia for the Study for Stung Chinit Water Resources
Development (TA 3275-CAM) in ADB. 1999. Monthly Report on Small-Scale Technical Assistance Projects Not
Exceeding $150,000 Per Project. Manila.
40 Appendix 7
ultimately laid down a critically flawed foundation for the project, which was to have long-lasting
ramifications affecting implementation and impact.
6. Although unknown at the time,7 two critical factors caused a number of oversights. The
first was that the small-scale TA failed to accurately identify the reality that much of the irrigation
infrastructure was in fact far beyond repair, and would require complete reconstruction. The
second oversight was that the irrigation design was based on what was later to be determined
as flawed topographic analysis. This had a cumulative impact of dramatically altering the costs
associated with reservoir construction and, accordingly, the economic feasibility of the project.
The flooding of the Stung Chinit River upstream from the weir site was also grossly
underestimated, and resulted in far greater resettlement than originally anticipated.
7. The consultants and ADB had some apparent differences in interpreting the available
evidence, for example, the consultants stated in their report that the weir gates and control and
diversion structures are no longer functioning, and some canals are filled with sediment or have
eroded banks, leaving many areas unsuitable for crops even in the wet season and entirely
barren during the dry season.
8. However, by the time the report and recommendation of the President was prepared
(2000), the view was more that the basic layout of the scheme is sound, the weir
superstructures are intact, and many of the canals require only moderate earthwork. The
situation presents an opportunity to restore irrigation at less cost than to establish a new
8. The loan for the project became effective in February 2001 and the project was divided
into four components, of which component 2 was classified as irrigation infrastructure,
comprising (i) repairing the main diversion weirs on the Stung Chinit river and its tributary the
Stung Tang Krasang; (ii) repairing flood bunds, constructing new bunds on the eastern side of
the main canal, and providing drainage structures to protect low-lying agricultural areas between
the two rivers; (iii) constructing a fish pass and riparian flow structures to allow annual fish
migrations; and (iv) remodeling and repairing the main and secondary canals and regulators,
along with roughly 60 kilometers of embankment service roads. The base cost for component 2
was estimated to be $5.98 million.
B. 2001–2003: The Initial Implementation Period
11. Implementation of the project commenced in September 2001 with the mobilization of
the Lahmeyer International and SMEC Joint Venture consultants.
12. In the progress report of September 2002 (i.e., well into the detailed design phase), the
situation had changed radically:
… on the basis of recent geotechnical and field investigations, that the structural
damage to the weirs at Stung Chinit and Stung Tang Krasang is so extensive that
demolition or abandonment of the existing structures, and their replacement by a
combination of “spillweir” and riparian outlet at each river, are recommended as being
more economical than repairing the original structures and supplementing them with
Possibility the project preparatory TA consultants did not fully appreciate the level of understanding of the ground
issues by local agencies at the time. .
Appendix 7 41
flood spillways. The Stung Tang Krasang weir shows similar signs of structural
13. The same report also contains a description of the hydraulic model studies of flood flows
through the existing and proposed new diversion structures at Stung Chinit and Stung Tang
Krasang; in short, the model was found to have been completely recalibrated and rerun once
the ground elevations taken from the orthophotomapping were found to be incorrect in many
C. 2003 Onward: Change of Scope, and Construction of Physical Works
14. In December 2002, a process of seeking alternative options was initiated, with
consultants advised to explore only those elements of the original project that could be
implemented within the existing original loan amount. The project implementation options report
was prepared in early 2003.9 It explored a series of 13 alternative options (Table A8). To
maintain costs within the original loan amount, a modular approach to project implementation
was recommended. By late 2003, the preferred project implementation arrangement was to
construct the core infrastructure on the Stung Chinit reservoir10 and in a priority area to the
north11 using only currently available funding, and to expand the scheme later.12
15. The irrigation command area under this scope change declined significantly. The new
configuration was to provide primary, secondary, tertiary, and quaternary13 infrastructure for
irrigation and drainage, in a priority area west of the northern main canal of about 2,960 ha
(compared with 7,000 ha envisaged at loan appraisal) and about 1,800 ha during the dry
season (compared with 2,000 ha envisaged originally). While the command area and
infrastructure requirements decreased significantly, cost implications as a result of the scope
change were unfavorable. The original component 2 base cost estimates increased from
5.9 million to 8.4 million.14
16. ADB approved the major change of scope in December 2003; nearly 2 years had
passed since the unknowns became known. The major impacts of this outcome were the
(i) Component 2 (irrigation infrastructure) was significantly delayed,15 while the
progress of component 1 (farmer community organization and extension
MOWRAM. 2002. Stung Chinit irrigation and Rural Infrastructure Project: Loan 1753 (SF). Monthly progress report,
page 10. Phnom Penh
In late 2002 a series of recommendations were made; several focused on the need to undertake extensive soil
tests to ascertain the long-term agricultural potential of the project area soils and their suitability for construction.
The issue of soil quality became especially relevant after the change of scope in late 2003.
With a reservoir upper storage level of 252.00 meters, the reservoir was expected to hold 3.56 gigaliters.
The southern irrigation command area and control structures across the Stung Tang Krasang were to be held over
for potential phase 2 developments. The irrigation off-take structure at the reservoir to the south of the Stung Chinit
was to be constructed in preparation for a potential phase 2 expansion into the southern command area (potential
area: 4,000 ha).
As and when funding might become available.
Quaternary canals were only constructed in the trial area of the irrigation system. The initial design was to
construct regimented consistent quaternary canals for ease of management, however land titling issues and loss of
irrigation area greatly impacted their adoption. Presently farmers are designing quaternary canals within their own
irrigation bays on an ad hoc basis.
These cost changes were borne by a reallocation of funds within the project (reducing the scope of rural
infrastructure investments) and the contribution of $3 million from MOWRAM.
The initial plan. which the other components were using, was to deliver water to farmers by late 2004. This was
now delayed to mid-2006 due to the change in scope.
42 Appendix 7
services) was hampered by the absence of irrigation infrastructure to perform
(ii) An experimental irrigation plot (about 50 ha) was established adjacent to the
northwestern end of the main canal and water was pumped from the nearby
original Stung Tang Krasang infrastructure. At this time the issue of soil quality
analysis was raised again, as sandy soils with high permeability rates were
present in the reduced command area.16 This represented a significant setback
for the potential of dry season rice production in the northern command area.17
(iii) A new resettlement plan had to be developed (as a result of the new dam level),
which significantly delayed progress.
17. The construction of infrastructure on the Stung Chinit reservoir then continued, including
the construction of primary, secondary, and tertiary canals and associated drainage.
D. 2008 and Onward: Project Completion
18. Many minor construction modifications were required post-2003 and a 1-year extension
on the completion date was necessitated (to June 2008). A description of the completed
irrigation infrastructure (based on data provided by MOWRAM and maintenance and completion
reports drafted by Lahmeyer–SMEC) is as follows: The irrigation scheme at this primary stage
comprises a reservoir with 750 m spillway (height 252.00) with an upper storage level volume at
this height of 3.56 gigaliters. A 120 m fish ladder plus 120 m riparian flow structure are present
adjacent to the northern end of the spillway. Head-works are in place to regulate water to the
North and for future canal expansion to the south of the Stung Chinit. Head-works also exist at
the O Chok reservoir to regulate water for future expansion in this area. Six outlet structures
have been constructed in the reservoir embankments allowing lands to be irrigated adjacent to
the reservoir.18 The main canal, some 9 km in length, is best described hydraulically as a series
of three smaller reservoirs divided in sections by a culvert and a siphon. The rate of irrigation
water distribution is determined by the water level of the reservoir via gate regulation influencing
the standing water level of the three sections of the main canal. Maintaining a constant height of
water in the main canal maintains consistent flow rates through the 5 secondary canals and 38
km of tertiary canals into 47 tertiary blocks with a total command area of 2,214 ha19 in the wet
season. The system has been designed so no gate management is required after the secondary
gate and water is distributed evenly across the command area by proportional flow.20 Drainage
is provided by means of 5 secondary and 49 tertiary drains with a combined length of
approximately 70 km.
This is a setback for the production of dry season irrigation as high percolation rates make maintaining appropriate
irrigation water levels within a specific plot almost impossible. The farmer water user community (FWUC) stated
during field interviews for the project completion report preparation that 30% of the northern command area
comprises sandy higher percolation soils. The 2008 completion report by Lahmeyer–SMEC states that while soils
of higher permeability exist, the general clay content is sufficient for rice cultivation if managed appropriately. The
system is designed to deliver 117 millimeters per week to the field at full supply discharge, which is more than
adequate if water is not wasted.
Other crops could potentially be produced on these soil types such as peanuts and vegetables.
The consultant was unable to gain access to these sites to confirm status during the field visit (September 2009).
This figure is from the 2007 Lahmeyer–SMEC Irrigation System Management document and differs from the more
usually quoted figure of 3,000 ha. In its project completion report, MOWRAM states that 1,973 ha are irrigated off
the main northern canal, although the area collected under an irrigation service fee by the Stung Chinit FWUC in
2008 was 2,398 ha. The original command area serviced at the change of scope was to be 2,960 ha. Overall, the
size of the dry and wet season planted areas at the completion of the project cannot be stated with a high degree
While this makes water management simpler for primary and secondary canal management, it requires greater
coordination at field level to ensure correct flow rates are maintained 24 hours per day.
Appendix 7 43
19. A positive consequence of the change of scope and modular approach within the
irrigation infrastructure component is the potential for cost-effective expansion, since the sunken
costs were already absorbed by the successful completion of the reservoir and associated
offtake structures. The 2007 Irrigation System Management Report produced by Lahmeyer–
SMEC states that the recently constructed infrastructure offers the potential for extending the
irrigation command area without major additional infrastructure costs. A further 11,000 ha has
been identified for potential expansion, comprising the Southern Irrigation Area, 7,000 ha; the
Banteum Reach Village, 700 ha; areas with recession cropping in the reservoir, 550 ha;21 areas
adjacent to the Chinit reservoir, 1,000 ha; the O’chok area, 750 ha; and the Lok Tah and Kai
Tah reservoirs command area, 1,000 ha.
20. The main points emerging from this history of the project’s evolution are as follows:
(i) The decision not to proceed with an upstream reservoir for irrigating some
12,000 ha in the Stung Chinit catchment appears to have been appropriate.
(ii) Unfortunately the irrigation design relating to a storage weir across the Stung
Chinit and Stung Trang Kasang—with an associated command area of
7,000 ha—was overly ambitious and flawed from the commencement of the
project, due largely to inaccurate data sets and analyses, combined with the
inability to assess the condition of the structures requiring rehabilitation. This led
to serious ramifications on the project’s cost structure, the extent and complexity
of the resettlement plan, and the length of time required to reassess operating
conditions and to devise a change of scope for the construction of the new
(iii) Further, the absence of sufficient analysis of the soil structure for agricultural
suitability (despite this being advised in 2003) did even more to set back and
undermine confidence in the project’s management.22
(iv) The decision to modulate development at the time of the scope change was a
positive approach to a difficult (if somewhat self-inflicted) predicament. The time
taken to devise and approve the change of scope had a negative effect on the
other project components, not only delaying overall progress but undermining
confidence in the project (within ADB and MOWRAM).
(v) While the project appeared to have great irrigation potential 12 years ago, and
the engineering standard of the current development is high, only a small degree
of that potential23 has so far been realized.
(vi) The current condition of existing irrigation infrastructure must be maintained to a
sufficient standard to realize the system’s expansion potential.24
During field interviews, people stated that currently no recession rice was being produced in the reservoir. This was
not verified by site inspection.
While the soil issue is important it is manageable in the future with appropriate agricultural practices and cropping;
however, these are more complex, foreign, and take time to learn.
Other benefits include fish production and rural infrastructure improvement; however, this paper relates only to
A proposal for urgent maintenance was drafted in mid-2008. During the project completion review site visit of
September 2009 no maintenance actions were noted. This would also bring overall average development costs
44 Appendix 7
Table A7: Change of Scope Options
Dry Dry Dry
season season season Potential
Base Benefits irrigation irrigation irrigation total dry
Costs wet from from from season IRR
($ Area season inflows storage storage irrigation Irrigation
Option Description million) (ha) (ha) (ha) TK (ha) CH (ha) (ha) only (%) Advantages Disadvantages
0 Original design and 6.0 7,00 7,000 2,000 0 0 2,000 15.0
original cost estimate 0
from ADB RRP
1 Stung TK + MCN only 5.5 2,96 2,960 200 0 0 200 Early generation CH flows not
with TK @min height 0 of irrigation mobilized and no
(no CH embs) benefits benefits South of
2 Stung TK + MCN only 5.7 2,96 2,960 200 480 0 680 Early generation CH flows not
with TK storage (no 0 of irrigation mobilized and no
CH embs) benefits and benefits South of
improved benefits Stung Chinit
3 Stung CH + MCN 7.5 2,96 2,960 1,800 0 0 1,800 CH flows Later start to
only @ min height (no 0 mobilized irrigation and no
TK embs) benefits South of
4 Stung CH + MCN 7.7 2,96 2,960 1,800 0 588 2,388 CH flows Later start to
only with storage (no 0 mobilized and irrigation and no
TK embs) improved benefits benefits South of
5 Stung TK + CH + 9.0 2,96 2,960 2,000 0 0 2,000 6.1 In line with No benefits South of
MCN only embs @ 0 staged planning Stung Chinit
6 Stung TK + CH + 9.5 2,96 2,960 2,000 480 588 3,067 Significant High storage not fully
MCN only embs with 0 increase in usable until later
storage benefits over stage of project
option 5 realized
7 Omission of SC7 and 11.9 4,95 4,953 2,000 0 0 2,000 Some benefits Full benefits south of
SC8; TK & CH embs 3 south of Stung Stung Chinit not
@ min ht. Chinit realized
8 Omission of SC7 and 12.4 4,95 4,953 2,000 480 588 3,067 8.6 Mobilization of Full benefits south of
SC8; TK & CH embs 3 active water Stung Chinit not
with storage storage realized
9 Stung CH only + 13.7 7,38 7,381 1,800 0 0 1,800 CH flows Later start to
MCN & MCS (no TK 1 mobilized and irrigation and TK
embs) @ min height benefits South of flows not mobilized
Appendix 7 45
10 Stung CH only + 14.0 7,38 7,381 1,800 0 588 2,388 CH flows and Later start to
MCN & MCS (no TK 1 active storage irrigation and TK
embs) with storage mobilized flows not mobilized
11 Full infrastructure @ 14.4 7,38 7,381 2,000 0 0 2,000 9.1 Original planning Low IRR from
a min height 2,000 ha 1 increase costs
11 Full infrastructure @ 15.2 7,38 7,381 2,000 0 0 2,000 8.6 Original plus Further reduction in
b min height 7,000 ha 1 tertiary works IRR
12 Full infrastructure with 15.7 7,38 7,381 2,000 480 588 3,067 11.0 Significant Higher costs
storage 1 increase in
Source: Ministry of Water Resources and Meteorology.
46 Appendix 8
REVIEW OF PROJECT ECONOMIC ANALYSES
A. Economic Analysis at Project Design
1. Project design was undertaken in two stages: (i) in the 1996 technical assistance
(TA) for Stung Chinit Water Resource Development Project, and (ii) the 1999 small-scale TA
for the Study for Stung Chinit Water Resources Development.1
2. The economic analysis under the first TA was cursory (as was the whole design
basis at that stage). Internal Asian Development Bank (ADB) review concluded that (i) the
economic pricing was largely incorrect; (ii) estimates of likely farm and crop financial and
economic budgets were entirely from secondary sources; (iii) projections of benefits from fish
ponds were overstated; (iii) numerous arithmetic and computational errors were overlooked;
and thus (iv) the economic analysis as a whole was flawed, incomplete, and inadequate.
3. However, this stage of economic analysis did highlight two important points: (i) the
associated rapid socioeconomic appraisal of the stung chinit basin paper did document the
range of nonfarm activities that predominated in the proposed project area (and which then
seemed to get overlooked in subsequent project design), and (ii) it concluded that a large-
scale irrigation project rehabilitating existing infrastructure was not viable without the
construction of a dam in surrounding hills. Such a dam was unacceptable for environmental
reasons and also because the area was still controlled by the Khmer Rouge.
4. The subsequent 1999 small-scale TA report is the basis of the eventual project
viability estimate and is reflected in the economic analysis. Some main features of this stage
of the economic analysis are as follows:
(i) The economic analysis was restricted to a cost–benefit analysis, and wider
aspects (e.g., of economic rationale, alternatives analysis, sustainability of
impact) were not covered.
(ii) Incremental rice production was the biggest source of benefits, with modest
amounts of dry season cropping and only some small fisheries impacts (about
5% of total benefits).
(iii) The economic pricing was done properly, with rice economic valuation based
on a weighted import and export parity price basis (which was reasonable at
that time—given that Cambodia was importing food).
(iv) The project now included a rural infrastructure component (i.e., part B); the
estimated economic internal rate of return of this component (32.4%) was
more than double that of the irrigation scheme alone (15%).
(v) Perhaps more importantly, the economic analysis of the project as designed
failed to provide any real technical or economic rationale for the inclusion of
this rural infrastructure component, other than in a very general way of
providing links from the irrigation scheme to local markets. The clear (if crude)
implication is that part B of the project was grafted on to the former irrigation
scheme design simply to raise the overall estimate of project viability above
the 12% threshold and to increase the size of the eventual investment project.
(vi) The estimation of benefits for the rural infrastructure component included not
only vehicle operating cost savings but also an estimated impact on
agriculture producer surplus. The description of such benefits does not accord
with what is usually measured formally as producer surplus (which is the
ADB. 1996. Technical Assistance to the Kingdom of Cambodia for the Stung Chinit Water Resource
Development Project. Manila (TA 2592–CAM); Small-scale technical assistance to the Kingdom of Cambodia
for the Study for Stung Chinit Water Resources Development (TA 3275–CAM) in ADB. 1999. Monthly Report
on Small-Scale Technical Assistance Projects Not Exceeding $150,000 Per Project. Manila.
Appendix 8 47
difference between the price some producers actually do sell at compared
with a market clearing price), and whether or not project farmers get such
benefits depends upon the competitiveness of input and output markets
(including local transport services). In other words, the methodology was not
entirely appropriate, and really tried to describe dynamic impacts that might
result from lower vehicle operating costs in the longer term in the area around
the irrigation itself.2
(vii) Yet more serious is the omission of any significant analysis of risk and
uncertainty in conditions where it must have been quite obvious how modest
the overall information base was, and how much was not known about the
project’s operating environment. Simple increases or decreases in aggregate
benefit and cost flows (plus implementation delays) were all that were
investigated through sensitivity testing, even though (for example) risk
modeling software was available and being used in other sectors (e.g.,
power) in ADB at the time.
6. In sum, the economic analysis was used almost entirely as a way of justifying the
proposed project rather than as an investigative and design tool.
B. Economic Analysis at Change of Scope
7. At the change of scope in 2003, the revised economic analysis centered on the
preferred design option of reservoir construction, main canal reconstruction, and canal
network installation, on the basis of with- and without-sunk costs situation (the sunk costs
being those already incurred by implementing the project to that point).
8. Some of the main analytical features of what was an extensive and comprehensive
approach include the following:
(i) The economic pricing assumptions were repeated, although rice was now
valued on an export parity pricing basis (and thus its economic value was
lower than at appraisal).
(ii) The reservoir construction created an estimated stream of fisheries benefits—
at almost half the scale of those from incremental rice production.
(iii) Economic losses due to resettlement and/or land loss were identified.
(iv) The agriculture producer surplus methodology was repeated for rural road
(v) A number of nonquantified benefits were described but not calculated (e.g.,
improved access), although some double counting (e.g., health benefits from
increased fish consumption) are mentioned, but would already have been
included in the economic valuation of the fish prices reflecting consumers’
rational awareness of all benefits.
(vi) Benefits from time savings and reduced spoilage at markets were now
included in part B benefits (they had not earlier been calculated).
(vii) Poverty impact was estimated, although it is not clear if this is the standard
poverty impact ratio as usually estimated for ADB projects.
(viii) The analysis of unknown outcomes was again restricted to very simplified
sensitivity testing (including calculating switching values for a few variables),
despite the fact that the seriousness of the lack of real knowledge about
conditions in the project area must have become quite obvious, and tools for
It may have been reasonable to qualitatively describe such expected impacts, but to also calculate them in
relation to a without-project scenario in circumstance when project implementation would be illogical.
48 Appendix 8
risk analysis were being routinely and formally recommended for use within
9. Of these features of economic analysis, the one that is perhaps most glaring is that
the analysis of risk was not undertaken at all. A few specific uncertain states and outcomes
may have been described through sensitivity testing, but a more probability-based analysis
centering on a few key variables’ distributions (e.g., scale of scheme development, wet and
dry season mixtures and yields, price of rice, and O&M costs) could have provided
considerable insight into the real sources of concern about the project.
C. Economic Viability at Project Completion
10. Ideally, estimation of economic viability at project completion would be based on data
and findings from impact evaluation surveys directly linked to project interventions and
11. In fact, partly because the project framework was so weak, the means of verification
for impact indicator measurement were extremely loosely specified (i.e., progress reports,
review missions, benefit monitoring and evaluation surveys repeated for outputs, purpose,
and goals), and at the time of preparing the project completion report (September 2009) no
rigorous or systematic-collected impact data exists upon which to reestimate project viability.
12. To assess the effectiveness of outcome achievement and the efficiency of
achievement of outputs and outcome, the following types of data would be needed:
(i) agronomic and yield data from crop-cutting on a stratified (e.g., location, crop, variety,
season) sample of irrigation blocks; (ii) crop budget and farm income calculations, based on
yields data, cropping patterns, and any project-induced changes thereof; (iii) income and
expenditure data (again, stratified by location, family poverty status, and household
characteristics); (iv) vehicle operating costs (by type of vehicle) and passenger time savings;
and (v) market wastage and spoilage estimation from traders, wholesalers, and processors.
In addition to quantitative estimates of direct project impact, this would need to be compared
with nonproject (i.e., control group) populations and against the context of other influences
and factors that may have contributed to observed outcomes (e.g., other projects in the area,
government sector policies, more general macroeconomic conditions affecting employment,
13. A comprehensive impact analysis of the project on the basis of such information and
including a reestimation of the project’s economic internal rate of return is still possible—and
is especially relevant given the delay in physical works construction (being completed only in
2008) and the small number of seasons so far for joint farmer and agency cooperation to
embed itself into stable scheme management. An appropriate time for such work (e.g., in a
performance evaluation report) would be 5 years after completion of construction, i.e., in
2013. Given that the project has proved so different in implementation to what was originally
expected at appraisal (already a decade or more prior to the project completion report), any
estimation of economic viability and explanation of reasons for such is likely to be most
14. Given that the kind of formally derived, quantitative information as just described is
not available, a qualitative assessment of the situation vis-à-vis scale of economic impacts at
project completion suggests the following:
(i) Available information from several sources on wet season rice yields
suggests a significant productivity and aggregate production impact already.
For example ADB. 2002. Handbook for Integrating Risk Analysis in the Economic Analysis of Projects. Manila.
Appendix 8 49
Small-scale, informally sampled crop-cutting exercises in 2007 and 2008 wet
seasons organized by Groupe de Recherche et d’Echanges Technologiques
(GRET) and the farmer water user community (FWUC) suggest that average
yields across the scheme may have risen to some 2.2–2.3 tons of paddy per
ha, up from 0.8–1.0 tons prior to the scheme construction. Triangulating this
estimate via on-site discussions with individual farmers and FWUC
representatives (including some based on traditional tang4 estimates of
output) confirm the scale of this increase, i.e., of about 30%–40% increase in
average productivity. Yields of around 3 tons/ha are now not uncommon
(especially in the more intensively transplanted regimes on the south of the
scheme—around SC4 and SC5), although in the sandy areas (to the north—
SC1 and SC2, where more broadcasting of seed occurs) 1.3 tons/ha is also
found. A slight increase of about 5% in overall average yield was estimated
between 2007 and 2008 wet seasons, and the FWUC reports less wet season
rice yield variability across plots and farms and over time than previously.
(ii) The implication of this kind of average area increase is that something over
2,000 ha5 of wet season rice are now each producing about 1.3–1.5 extra
tons of paddy than before the scheme construction. This probably represents
about 2,000 tons of milled rice (based on a milling percentage of 65%). With
the current (late 2009) financial price of milled Thai 5% broken rice FOB (free
on board) Bangkok being around $500 per ton, an economic price of rice for
Cambodia based on quality adjusted export parity and calculated at the
project site value (i.e., net of transport costs to Sisowath Quay and Bangkok)
is around $450 per ton. This implies annual gross benefits from wet season
rice cropping from the project at around $0.9 million a year. As the
incremental expenses on fertilizer (DAP-16-20-0) are typically only modest
(and labor requirements—especially for weeding—are probably reduced, the
annual net economic benefit flow may therefore be around $0.75 million.
Given that average yields are expected to increase significantly in the future,
this benefits stream should change commensurately, although it will of course
be affected by world price developments.6
(iii) Supplementary information from surveys7 suggests that livelihood impacts
include physical and financial asset bases, food security, and expectations
about the future. These benefits are recorded to be highest among poorer
families in villages around the irrigation scheme, i.e., those families most
dependent upon agriculture as compared with those with significant
nonfarming interests. The survey also suggests great willingness for further
agriculture-based livelihood improvements given that food security is now
assured, based on activities such as vegetables and pig-raising. (This may be
especially true for farmers on the sandier northern soils and for those families
with fewer nonfarm opportunities.) The survey data suggest that the biggest
income impacts have been on the lives of the poorest quartile, and that nearly
all respondents were more optimistic about the future than before the
scheme’s construction. Visual evidence of increased building in the scheme
vicinity and of more motorcycle ownership abounds, although some of this is
of course related to more generalized macroeconomic improvements.
A Khmer measure, 1 tang equals 24 kilograms
Some small areas on the scheme remain unplanted and one village only recently joined the scheme.
Since the project was appraised, the real price of rice has more than trebled (2008) and is now more than
double what was anticipated at the change of scope analysis.
Centre d'Etude et de Développemetn Agricole Cambodgien (CEDAC). 2007. Toward Livelihood
Improvement—A Project Impact Evaluation Report. Phnom Penh.
50 Appendix 8
(iv) Since the project was implemented the value of land on the scheme has
increased significantly; land which used to sell for $150–$200 per ha in the
late 1990s was selling for almost 10 times this figure at the height of the
recent boom in 2007. Realizable land values are probably best estimated at
several times prescheme value, and at least some part of this is due to
agriculture productivity increase and production potential.
(v) Little evidence can be found that that scale of fisheries benefits approaches
what was estimated at change of scope. Some fishing is ongoing from the
reservoir, but as nearby rivers are canals are also a source of abundant fish-
based protein, no formal or informal evidence for identifiable incremental
harvest and/or production from the project investment exists, although there
may be an as yet unidentified stock impact.
(vi) No formal evidence for vehicle operating costs is available, but local transport
fares over project roads have decreased substantially. For example, transport
to local markets (e.g., in Kampong Thmor) over 20 km, which used to cost KR
9,000 per person now costs around KR 2,000. Journeys that used to only be
taken by motorbike are now covered by small taxis and tuk-tuks, and traffic
volume on some project roads (e.g., Kvek to Dong Kola) is estimated by the
Provincial Department of Rural Development to have increased tenfold. (In
areas linked to new rubber development, nearby traffic volume has doubled
(vii) No evidence is available regarding market losses and spoilage avoided, but
market fee revenues are rising, and the numbers of sellers and periods of
activity have both increased at the new market sites (covered and
uncovered), indicating an increase in overall economic activity and
fundamental viability and likely sustainability.
(viii) Intangible benefits are evident in terms of capacity development (e.g., of the
FWUC), almost certainly health benefits have resulted from the reduction in
pesticide use, general access to social (health and education) services in the
area has improved, and forward and backward agriculture linkages have been
strengthened as marketed volumes have increased.
(ix) A major latent benefit of the scheme construction so far is that its expansion
(at least a further 11,000 ha—based on the current reservoir capacity) to (for
example) the south—and as was originally intended at appraisal—can now be
achieved at low marginal cost. There is scope for private (foreign, if
necessary) investment to utilize the presently unused reservoir capacity and
pay a fee to the state for such service, thereby potentially capturing a benefit
flow for current scheme operating support.8
D. Summary Analysis Findings
15. The main conclusions from this review of the project’s economic analysis are
therefore as follows:
Indeed, some draining of the reservoir is already reported to areas other than those covered by project; if so,
this represents a stream of benefits occurring to nonproject sites and should be investigated during a
performance evaluation report exercise.
Appendix 8 51
(i) Economic analysis was taken to be cost–benefit estimation alone, and
neglected wider economic considerations.
(ii) The emphasis of the economics work was primarily on justifying particular
investment proposals rather than using analytic tools for in-depth investigation
and formulation of alternatives.
(iii) Much more use could have been made of risk analysis tools in a project
preparation and implementation situation that was particularly prone to lack of
(iv) Reestimation of the project’s impact and viability is still possible, provided a
range of data is collected in a timely and methodological fashion after the
project evaluation report stage.
(v) Relatively informal evidence suggests that economic benefits from rice
production are already substantial, and the 40% command area decline (vis-
à-vis appraisal) has been partly offset by the real economic rice price increase
in recent years. Benefits from fisheries may have been overestimated, and a
large range of nonquantifiable benefits which the investment has clearly
(vi) The economics of the investment under the project would change significantly
if the scheme were to be expanded on the basis of the newly provided
reservoir capacity (i.e., its embedded sunk costs mean that the costs of extra
canals will be relatively minor).
52 Appendix 9
1. The eventual economic viability of the project is a function of its relatively long-term
sustainability. This is particularly so because its development cost per hectare of irrigation
area commanded has been much higher than expected at appraisal (and thus at completion
the investment looks inefficient), yet a clear benefit stream is identifiable from several
sources and this has the capacity to increase significantly in future, but only if certain
conditions are met. The conditions determining sustainability can be divided into those that
are within the control of project management to affect and those that are not.
A. Nonproject Circumstances Affecting Sustainability
4. Some of the issues surrounding conditions affecting project sustainability that are
outside the scope of project management include the following:
(i) Commodity price developments—the extent to which the scale of economic
benefits (and financial returns to farmers) are maintained—will depend
significantly on commodity price (and especially rice price) developments.
These are unpredictable and variable; since the project was first considered
the real price of rice rose almost fourfold and then fell back to about double its
appraisal price. Current projections are for real prices to fall somewhat in
coming years, but given the present extent of international interest in
developing national and private supplies from specific sites this could change.
(ii) The competition from nonagriculture activities (e.g., construction, trading, and
transportation) will determine how many farmers will devote time and
resources to rice-based and other farming activities. The economic benefits
stream could well diminish if broader economic opportunities intrude upon
agriculture’s relative attractiveness.
(iii) The benefits stream will be sensitive to climate impacts, and specifically to
changes in climate patterns. It may be expected that in conditions of both
increasing drought and increasing floods, which may be anticipated in future
years, improved water management (i.e., storage and drainage) will be
beneficial to productivity and production.
(iv) The wider macro sector (i.e., agriculture) and subsector (i.e., water resources
and irrigation) policy environment will also affect how easy it is for farmers to
make and retain profits from agriculture activities, to gain access to credit,
and for irrigation management transfer to be possible and facilitated.
Obviously, the legislation and policies surrounding water resources (e.g., the
Law on Water Resources Management, 2007) and irrigation (e.g., Circular
No. 1 of 1999) will create or impede conditions for more participatory
management and irrigation management transfer.
B. Project Circumstances Affecting Sustainability
5. The circumstances within the control of project management that affect sustainability
are of most concern for future decision-making. Benefit sustainability depends in the first
instance upon maintaining the investment in the physical infrastructure of the project.
6. Regarding the physical infrastructure itself, the key structures requiring inspection
and maintenance (both annual and routine) are (i) the integrity of the reservoir embankments
(particularly the high erosion zones adjacent to control structures); (ii) the degree of rock
protection on the weir spillway and the embankment integrity; and (iii) high erosion zones
adjacent to control structures on the main northern canal, particularly around the secondary
offtake gates and siphon. A detailed operation and maintenance (O&M) manual was
prepared by Lahmeyer International– SMEC Joint Venture (SMEC).
Appendix 9 53
7. Implementing an O&M plan for the irrigation scheme requires (i) clear delineation of
responsibilities between the managing partners (i.e., state agencies and farmer
communities); (ii) each partner to have the technical capability to undertake the assigned
responsibilities; and (iii) each partner to have sufficient financial resources to allocate to
8. For the definition and delineation of respective O&M responsibilities, the letter of
agreement between Kampong Thom Provincial Department of Water Resources and
Meteorology (PDWRAM) and farmer water user community (FWUC) comprehensively
defines the management tasks by partner and type of infrastructure (i.e., headworks,
reservoir, gates, access roads, and various types of canals and drains). The definition of
responsibilities along these lines has taken the whole period of project implementation to
finalize—mainly because of the limited time since physical works completion and the need to
adequately strengthen the FWUC for water management operation. Now the second aspect
of O&M responsibility definition (i.e., the technical capacity to actually undertake assigned
tasks) has been achieved.
9. Where problems regarding project sustainability arise is with respect to the financial
aspects of O&M. On the part of state agencies (i.e., MOWRAM, PDWRAM) the question is
not of resources but rather of the actual commitment of adequate resources to project O&M
needs on a regular basis. Experience in Cambodia (and elsewhere in the region) suggests
that the culture of O&M is not fully institutionalized, and that routine maintenance is
especially neglected, resulting in greater (and less economic) periodic maintenance
investments and (more usually) even greater costs incurred in rehabilitation and
reconstruction. For the FWUC the issues are more complex, as the amount of resources
raised from irrigation service fees (ISF) do not fully cover O&M obligations and a shortfall
must be met from external sources (i.e., state agencies, development partners).
10. Currently, the broad features of the financial dimensions of O&M (as embodied in the
current letter of agreement) for the project seem to be as follows:
(i) PDWRAM O&M responsibility costs are estimated at about $183,000 per year,
those of the FWUC at $87,500 per year;
(ii) the FWUC operating costs are presently about $28,100 a year;
(iii) ISF revenues are about $10,000 a year at present, and are projected to rise to
about treble this by 2013–2014;
(iv) ISF rates are about KR20,000 per ha at present (about $5 per ha), and will rise to
about $15 per ha in 2013–2014;
(v) ISF collection rates are high in terms of farmers making payments (over 90%)
and total amounts paid (over 85%); therefore
(vi) the FWUC inevitably faces a large financial shortfall in future years, as the
revenues from ISF paid by the project farmers will only just about cover FWUC’s
operating costs and not the real costs of physical O&M of irrigation structures.
11. Determining how this situation will play out in coming years is not possible as (i) the
policy environment is evolving; (ii) farmers’ willingness to pay higher ISF may increase with
several years of success; and (iii) profitable cropping, and commitments from other
development partner and interested parties may be forthcoming. That the situation is
particularly fluid is partly due to the project implementation delays, which meant that the time
before project closing for the embedding of irrigation system management was very limited.
MOWRAM and PDWRAM support for the FWUC will effectively be a matter of policy choice;
however, what is now needed as a matter of some urgency to ensure project sustainability
and maintenance of benefits streams is clarity surrounding O&M financial arrangements.