Independent Evaluation - Home > Search

Implementation Completion Report (ICR) Review - Irrigated Land & Watershed Management

1. Project Data:   
ICR Review Date Posted:
Project Name:
Irrigated Land & Watershed Management
Project Costs(US $M)
 43.2  27.1
Dominican Republic
Loan/Credit (US $M)
 28.0  21.7
Sector, Major Sect.:
Agricultural extension and research, Irrigation and drainage, Central government administration,
Agriculture fishing and forestry; Agriculture fishing and forestry; Law and justice and public administration
Cofinancing (US $M)
 None  None
L/C Number:
Board Approval (FY)
Partners involved
Closing Date
12/31/2003 12/30/2004
Evaluator: Panel Reviewer: Group Manager: Group:  
Keith Robert A. Oblitas
Christopher D. Gerrard Alain A. Barbu OEDSG

2. Project Objectives and Components:

a. Objectives
(i) Improve income of the farmers in the project area.

(ii) Test and develop methodologies that can be replicated nationwide both for efficient production support services, water management and operation and maintenance of irrigation systems and for sustainable and environmentally sound watershed management.
(iii) Help reduce the fiscal costs of irrigation schemes through the adoption of appropriate cost recovery policies.

b. Components (or Key Conditions in the case of Adjustment Loans):
(i) Irrigation Infrastructure (Planned project costs US$20.2 million, Actual costs US$5.9 million) to rehabilitate, complete and modernize existing schemes and land levelling and drainage.
(ii) Nizao watershed management (Planned project costs US$2.9 million, Actual costs US$2.6 million) to pilot improved irrigation management systems through a participatory process for later replication
(iii) Agricultural development (Planned project costs US$5.0 million, Actual costs US$5.3 million) to provide agricultural support services, on-farm research and diffusion of improved technologies.
(iv) Institutional development (Planned costs US$4.0 million, Actual costs US$5.6 million) to build capacity of farmers' water user associations, facilitate access to credit and land titling, and support the implementation agency: the National Institute of Water Resources (INDRHI).
(v) Studies (No separate costing at appraisal, Actual project costs US$ 4.5 million) primarily on technical aspects of irrigation.
(vi) Monitoring and evaluation (No separate costing at appraisal, Actual project costs US$1.2) to develop management information systems for INDRHI and water user associations.

c. Comments on Project Cost, Financing, Borrower Contribution, and Dates
The project was characterized by extremely lengthy preparation and implementation: the gap between the PCD and Appraisal was nearly 5 years, between Appraisal and Approval 2 years, Approval and Effectiveness 2 years, and Approval to closure nearly 9 years.
Project preparation costs (identification through project approval) at US$190,000 were low. Also considering the long duration of preparation, which would tend to add to costs, the project's poor quality of entry may have been partly due to the limited resources used for preparation.
Two-thirds of the loan was utilized over the project's nine-year implementation period. The reduced project costs and loan utilization were primarily due to shortfall in the project's largest component; irrigation infrastructure. US$1.2 million of the infrastructure part of the project went, in agreement with the Bank, to finance emergency works after a hurricane. Just before closure the project was amended to include reimbursement of PPFs of US$2.2 million for the Water and Sanitation Management Project in Tourism Areas. This was done due to the country's difficult financial situation.

3. Relevance of Objectives & Design:

The overall relevance of project objectives was Modest.
Objectives: Project objectives were relevant to the 1995 CAS which emphasized growth, poverty alleviation and natural resources management. Government's interest in increasing the private sector in the economy was also supported by the project's irrigation management turnover activities. Growth was a particular problem; GDP was growing more slowly than population growth, and agriculture, supporting one third of the labor force, had a GDP growth half that for the economy as a whole. Nearly 60 percent of the farmers in the irrigated areas had marginal sized holdings (less than 1.5 hectares). Also, while no data is available, general experience is that watershed management activities tend to involve the poor more predominantly. Piloting catchment management activities would contribute to environmental stability. Improving water charge revenues would benefit the country's difficult fiscal situation. Thus, the project would directly contribute to all three CAS objectives. If successful, replication would have important national impact.

Design: In principle, the project's design supported the CAS and project objectives. Irrigation improvement linked with enhanced agricultural practices, good water user associations (WUAs) and cost recovery is a typical package for irrigated agriculture. The watershed package also followed good practice experience in other countries. The project design's promotion of new technologies, capacity building for the WUAs and implementing agency, research, and M&E are desirable concomitants.

However, the above "generic" suitability did not take account of the difficulties of working in the Dominican Republic, especially its rural sector, and the accumulated experience of the Bank in previous operations. The ICR comments on the problems of the Dominican Republic project portfolio over the 1970s and 1980s. OED's Country Assistance Evaluation of the Dominican Republic (2003) comments that the Bank has not had an effective dialogue or lending in the rural sector. The project implementing agency, INDHRI, the country's National Institute of Water Resources, had performed badly under previous Bank projects. There was little experience with water user associations and there was no clear agreement with Government on water charges. Moreover, the project did not take sufficient account of previous experience. Thus, as commented on in the ICR, the Project Completion Report (FY89) and OED Performance Audit Report (FY93) of the Nizoa Irrigation Project discussed issues that resurfaced in the Irrigated Land and Watershed Management Project: effectiveness delays, counterpart funds problems, non-compliance with legal covenants, weak performance of INDHRI and uncompleted infrastructure works.

The needed stage of preparation at Approval also needs to be assessed during preparation, and achieved before implementation. Given the Dominican Republic's limited capacity, a highly advanced readiness would have been appropriate, and preferably with key activities already under implementation. An option would have been to have a small pilot project before a larger project such as this one.

4. Achievement of Objectives (Efficacy) :

(i) Improve income of the farmers in the project area. Modest
Project impact is difficult to assess as data is limited. Small sample surveys at the end of the project found that farmers were very satisfied with both the watershed management and agricultural development components, and felt their incomes had increased. The ICR cites a more than 15% improvement in farm incomes, which would have been due to the agricultural development and watershed management components. Significantly higher impact could be expected when the irrigation infrastructure is completed (none of the irrigation infrastructure was completed at project closure).
(ii) Test and develop replicable methodologies for efficient production support services, water management, irrigation operations and maintenance and environmentally sound watershed management: Modest.
There was no completed irrigation infrastructure and thus no basis for assessment of this component. The agricultural development and watershed management components had successful results and farmers reported their satisfaction in an end of project survey.
(iii) Help reduce the fiscal costs of irrigation schemes through the adoption of appropriate cost recovery policies. Negligible. At project closure water tariffs in real terms were about the same as before the project.

5. Efficiency:

On the grounds that none of the irrigation infrastructure was complete at project closure, no ERR was calculated for the ICR. However, costs and time durations indicate major inefficiencies. As stated, there was zero completed achievement for the project's largest component, irrigation, despite a project implementation period of over 9 years (Board to project closure). Engineering design studies for irrigation works cost US$2.5 million and were only available in the later years of the project. The study cost was 42% of the cost of works undertaken during the project period.
6. M&E Design, Implementation, & Utilization:

Design: The Monitoring and Evaluation component was not specifically costed at Appraisal. Actual costs were US$1.2 million. The M&E component was intended to prepare and implement a project management information system and an environmental monitoring system, and to furnish regular management reports. The appraisal report lacks a log-frame, and the performance indicators are of physical targets and not outcome related.
Implementation: Neither the MIS nor the environmental monitoring systems were implemented. On reporting, the ICR comments that "almost no annual progress reports or annual project implementation plans were sent to the Bank." However, information cited in the ICR indicates that some data was compiled.
Utilization: Due to the M&E component's design and implementation problems, there was limited data to use. One very valuable action was a small survey of farmers done at project completion which assessed the degree of enthusiasm for particular components (especially the watershed and agricultural development components), income changes, and other more outcome-oriented information.
7. Other (Safeguards, Fiduciary, Unintended Impacts--Positive & Negative):

The project's two largely successful components - watershed management and agricultural development - were both implemented/assisted by non-government agencies; a dynamic local NGO for the watershed program and Utah State University for agricultural development. Performance of these two NGOs was much better than Government performance. Given the Dominican Republic's weak government institutions, outsourcing, which was not explicitly intended at project commencement, may be a direction for Government to increasingly take.
  • No safeguard issues are referred to in the ICR. Near the beginning of the project, unfamiliarity with the Bank's procurement procedures resulted in declaration of misprocurement for US$300,000 of expenditure.

8. Ratings:
OED Review
Reason for Disagreement/Comments
UnsatisfactoryUnsatisfactoryThe good performance of the watershed management and agricultural development components partly compensate for major weaknesses in other respects. Otherwise, a Highly Unsatisfactory rating would have been more appropriate than the Unsatisfactory rating here. A similar comment could be made regarding Bank and borrower performance.
Institutional Dev.: 
Bank Perf.: 
Borrower Perf.: 
Quality of ICR: 

- When insufficient information is provided by the Bank for OED to arrive at a clear rating, OED will downgrade the relevant ratings as warranted beginning July 1, 2006.
- ICR rating values flagged with ' * ' don't comply with OP/BP 13.55, but are listed for completeness.

9. Lessons:

Maximize achievement of key borrower actions during project preparations: Actions essential for project implementation and success, or which demonstrate borrower commitment, are best required before project implementation and as early as possible in the project preparation process. For instance; delayed design and bidding caused most irrigation works to begin only in 2003, eight years after project Approval. There was also no progress during project preparation on tariff increases, and none during project implementation.

Assess and ensure borrower commitment and political/social feasibility for implementing key actions: Reducing fiscal costs (cost recovery) was one of the three project objectives. There was no Government commitment to increase water charges, and in real value terms no progress was made. Also, the implementing agency (INDRHI) was not committed to participatory implementation of irrigation works.
Expeditiously close or restructure non-performing projects: The various lengthy time lags between PCD to Appraisal (5 years), appraisal to Board (2 years), and Board to Effectiveness (2 years), and the state of unreadiness of the project and lack of commitment of Government cited in the ICR, would suggest ample opportunities for taking stock of achievement and prospects and closing project preparation if appropriate. Within the first two years of the project's 9 year implementation (from Board to closure), the project's non-performance would also suggest closure or radical restructuring. On restructuring, one option might have been a revised project with only watershed management and agricultural development (both successful) and a very small pilot irrigation component linked with a water user association. This would have had a positive outcome and provided the experience for further replication.
Tailor project design to country experience, borrower capacity, borrower commitment and borrower experience with project components: As discussed in Section 3, while "generic" project objectives and design were relevant, when related to the Bank's experience in the Dominican Republic, the detailed project design confronted previous unsuccessful experience on many accounts, amongst them: poor country experience, very weak capacity of the implementing agency, poor performance of a previous irrigation project, minimal experience with water user associations, counterpart funds problems, weak compliance with covenants and slow construction of infrastructure. Given this, if there was to be a project at all, a very simple and limited design would have been more suitable.

10. Assessment Recommended?  No


11. Comments on Quality of ICR:

Satisfactory. The ICR is candid in evaluating the project. The structure is methodical with systematic discussion by the separate Objectives and Components. Generally, most discussion is supported by data. Improvements could have included: (i) the Lessons section could have grouped or structured the 10 recommendations to highlight the most important; (ii) the reasons for the much more positive assessment of project performance made by the borrower could usefully have been discussed; and (iii) more comprehensive discussion could have been included on the watershed management and agricultural development components, as the reasons for their success would provide useful guidance for further replication (or, alternatively, a separate report would serve the same purpose).

© 2012 The World Bank Group, All Rights Reserved. Terms and Conditions