3.1 Overall Considerations on Irrigation Management Research
3.2 Observations and Recommendations Made in Previous Reviews
3.3 Review of the Research Programme
3.4 Overall Assessment of Research and Recommendations
The 1994-98 Medium-Term Plan reorganized IIMI's research into five programmes - assessing and improving the performance of irrigated agriculture, sector-level management of irrigated agriculture, improving public irrigation organizations, toward the local management of irrigation systems, and sustainable management of water delivery and disposal - with four cross-cutting themes - gender issues, health and environmental issues related to irrigation, choice and use of technology, and improving the global database on irrigated agriculture. Research achievements, since the establishment of IIMI in 1984, have been regrouped into these five programme areas. Country-specific research in the 14 countries where IIMI was or is present has in most cases been conducted without formal connection with the research programmes or the cross-cutting themes. That part of country research which is not directly identifiable with research programmes is reviewed in Chapter 4, as it in general relates more closely to institutional strengthening activities.
Irrigation management research has particular characteristics which determine the research strategy to be followed. The subject of inquiry is the performance of complex systems and organizations, so research generally needs to be conducted with operating agencies and in general through action research, in which researchers observe the impact on performance of changes in management practices. This implies making research supportive of development objectives, which complicates the research methodology. The approach, however, offers the advantage that the process of research itself provides institutional strengthening and training, and thus enhances the chances of adoption of the recommendations derived from the research. The fundamental linkages between the management of irrigated agriculture and the vital biophysical base - e.g., in hydrology, agronomy, soil erosion, and soil salinity - imply that the research is multidisciplinary and that data collection is difficult and must often be sustained over long periods of time. Despite the difficulty of this approach to irrigation management research, criteria are available to assess research performance (see Chapter 5).
IIMI's methodological approach to irrigation management research thus requires significant direct involvement with operating irrigation systems, in part because this is IIMI's source of information and its laboratory, in part because lack of local skills prevents effective delegation of field research to local agencies and professionals in most countries where it works, and in part because its mandate gives IIMI developmental goals that go beyond those of other CGIAR Centres. Strategic research may thus be achieved by drawing generic lessons from many field action research projects. For IIMI, this is both a unique source of advantage in research, and a source of difficulties in organizing strategic research. Success requires the formulation of rigorous hypotheses, systematic data collection, and the use of research methods that will subsequently allow comparative analysis. For this purpose, the use of relevant theory is essential. Success also requires that headquarters research programmes, country research, institutional strengthening activities, and assistance in the introduction of improved irrigation management and policy making practices be integrated into a coherent overall programme.
Strategic research may also be conducted by devising new approaches to answer management and policy questions which can be analyzed in a single location. In that sense, not all strategic research in irrigation management is based on inter-site comparisons and generalizations. This latter type of research can be implemented in locations without critical mass of research staff and without multi-site projects. This suggests a three-pronged geographical strategy, with concentration of a critical mass of research in a few selected countries, organization of multi-country projects targeted at well defined research questions, and single-site development of new scientific approaches.
3.2.1 1987 Review Mission
3.2.2 1990 External Programme and Management Review
Similar recommendations to those made by this Review Panel have already been made in previous external reviews. Our Report recognizes many of the same gaps in research performance which were noted before, suggesting that the response to the earlier recommendations may not have been effective (see Appendix VI). Some of these lingering observations and recommendations are the following:
· Lack of clarity and specificity of objectives. Need to make sharper choices of priorities, with critical mass of resources for each choice and location.· Problem of quality control: recommend use of an external panel of experts.
· Weak headquarters supervision of country projects.
· Need to make sharper choices of priorities with critical mass and sustained presence in the chosen sites, i.e., not overextending programmes either geographically or in size.· Need to strengthen quality control mechanisms: recommend expert reviews and use of special panels of experts.
· Need to increase communications with collaborating agencies and with local irrigation officials and policy makers.
· Need to consider the role and welfare of women.
· Need to be continually watchful of not following the path of a consulting firm.
3.3.1 Programme 1: Assessing and Improving the Performance of Irrigated Agriculture
3.3.2 Programme 2: Sector-Level Management of Irrigated Agriculture
3.3.3 Programme 3: Improving Public Irrigation Organizations
3.3.4 Programme 4: Toward the Local Management of Irrigation Systems
3.3.5 Programme 5: Sustainable Management of Water Delivery and Disposal
3.3.6 Cross-cutting Themes
IIMI's five research programmes are reviewed, as currently defined in the MTP. All core (unrestricted and restricted) funded projects are assigned to one of these five programmes. Their total 1994 budget is US$ 2,386,000. There are additionally 17 country projects with a complementary budget of US$ 2,887,000 and 7 Pakistan projects with a complementary budget of US$ 536,000 which are not directly identifiable with the research programmes, although some have direct linkages to one or more of the programmes. Table 3.1 gives the list of these projects with their sources and levels of funding, and the country location, and Table 3.2 gives staff allocation across programmes. In what follows, we have grouped publications in three categories: scholarly publications (journal articles and in-house monographs and research papers), proceedings (books of conference proceedings and chapters in proceedings), and in-house country and working papers.
(in 1994: 36.6 research staff, a core budget of US$ 630,000, and 7 core projects)
The main objective of this programme is to develop a conceptual framework and a set of indicators for assessing the performance of irrigated agriculture in terms of efficiency, equity, welfare, and sustainability. Work in this area has been done collaboratively with IFPRI for the design of indicators which address agricultural performance and socioeconomic effects, and with ICID, ILRI (International Institute for Land Reclamation and Improvement), and IHE (International Institute for Infrastructural, Hydraulic and Environmental Engineering) for the development of indicators which characterize hydraulic performance. The results have been published in 11 scholarly publications, 16 proceedings, and 14 country and working papers.
This research aims at providing irrigation managers and sector policy makers with tools to assess quantitatively the performance of irrigation systems and to monitor the impact on performance of changes in the water delivery system, the organizational structure, and the policy framework. It has done this by: (1) adapting indicators taken from other fields to the subject of water management in developing countries and systematizing them in a nested hierarchy of indicators, running from the irrigation system to the politico-economic system; (2) developing more systematic links between the physical and non-physical aspects which affect the performance of irrigated agriculture. A comparative analysis of performance indicators, using irrigation systems spread over different agroclimatic regions and management systems is in progress. Today, there are recorded examples of use of these performance indicators in several countries, among them Indonesia, Nepal, Sri Lanka, Pakistan, India, and Malaysia.
Table 3.1 1994 Budget Allocation and General Features by Research Themes and Country Projects
Programme 1: Assessing and Improving the Performance of Irrigated Agriculture
|
Project |
Status |
Budget |
Budget |
Budget |
Total |
Donor |
Country |
|
1. Performance - Coordination |
1 |
82,000 |
77,000 |
|
159,000 |
Japan Gov. |
|
|
2. Performance Assessment and Evaluation of Management Interventions - An Action Research Study for MUDA Irrigation Scheme in Malaysia |
1 |
85,000 |
88,000 |
|
173,000 |
Japan Gov. |
Malaysia |
|
3. Assessing and Improving Irrigation Management Performance in Selected Systems - Morocco |
1 |
105,000 |
|
|
105,000 |
Neth. Gov. |
Morocco |
|
4. Assessing and Improving Irrigation Management Performance in Selected Systems - Argentina |
1 |
26,000 |
45,000 |
|
71,000 |
Neth. Gov. |
Argentina |
|
5. Socio-Economic Impacts of Improved Irrigation Management Performance |
1 |
7,000 |
|
|
7,000 |
Japan Gov. |
Sri Lanka |
|
6. Testing Performance Assessment Indicators in Mahi-Kadana System - |
1 |
29,000 |
|
|
29,000 |
Japan Gov. |
India |
|
Gujarat, India |
1 |
13,000 |
|
|
13,000 |
Neth. Gov. |
India |
|
7. Current Practices in Irrigation Performance Assessment: A Review of Experiences in |
1 |
35,000 |
|
|
35,000 |
Japan Gov. |
Pakistan |
|
Pakistan |
1 |
15,000 |
23,000 |
|
38,000 |
Neth. Gov. |
Pakistan |
|
TOTAL |
|
397,000 |
233,000 |
0 |
630,000 |
|
|
Programme 3: Improving Public Irrigation Organization
|
Project |
Status |
Budget |
Budget |
Budget |
Total |
Donor |
Country |
|
|
|
|
|
|
|
|
|
|
1. Improving Public Irr. Organizations |
1 |
|
185,000 |
|
185,000 |
|
|
|
TOTAL |
|
0 |
185,000 |
0 |
185,000 |
|
|
Programme 4: Toward Local Management of Irrigation Systems
|
Project |
Status |
Budget |
Budget |
Budget |
Total |
Donor |
Country |
|
1. Potential for Water Markets Development in Pakistan |
1 |
|
30,000 |
|
30,000 |
|
Pakistan |
|
2. Gender issues and Irr. Management |
1 |
64,000 |
|
|
64,000 |
Neth. Gov. |
Sri Lanka/ Burkina Faso/Nepal |
|
3. Privatization and Self Management |
1 |
240,000 |
|
|
240,000 |
Germany |
|
|
4. Colombia Basin Follow-up |
1 |
5,000 |
|
|
5,000 |
Germany |
Colombia |
|
5. Participatory Management Study |
1 |
13,000 |
|
|
13,000 |
Germany |
Sri Lanka |
|
6. China Case Study - Changing Patterns of Irrigation Management in North China |
1 |
5,000 |
|
|
5,000 |
Germany |
China |
|
7. Irrigation Management Turnover in India |
1 |
64,000 |
|
|
64,000 |
Germany |
India |
|
8. Sudan Privatization Study |
1 |
20,000 |
|
|
20,000 |
Germany |
Sudan |
|
9. Irrigation Management Transfer Conference - Wuhan, China |
1 |
108,000 |
|
|
108,000 |
GER/FF |
China |
|
|
3 |
86,000 |
|
|
86,000 |
FF/FAO |
China |
|
10 Programme of Support Systems for Local Management |
3 |
118,000 |
|
|
118,000 |
Germany |
|
|
11 TARC/IIMI Studies at Thirappane |
1 |
|
20,000 |
|
20,000 |
|
Sri Lanka |
|
12 Latin America Programme - Phase II |
1 |
63,000 |
|
|
63,000 |
FF |
LAC |
|
13 Latin America Programme |
3 |
71,000 |
|
|
71,000 |
IDB |
LAC |
|
TOTAL |
|
857,000 |
50,000 |
0 |
907,000 |
|
|
* Compl. = Complementary
Status 3 grants are discounted by 33%
Status 4 grants are discounted by 50%Note: Programme 2 is not functional
Programme 5: Sustainable Management of Water Delivery and Disposal
|
Project |
Status |
Budget |
Budget |
Budget |
Total |
Donor |
Country |
|
1. Water Management Base Project |
1 |
|
60,000 |
|
60,000 |
|
|
|
2. Research on Irrigation and Drainage Technology |
1 |
60,000 |
44,000 |
|
104,000 |
Germany |
|
|
3. Research on Irrigation and Drainage Technology |
4 |
72,000 |
|
|
72,000 |
Germany |
|
|
4. Canal Operations |
1 |
|
43,000 |
|
43,000 |
|
Sri Lanka |
|
5. Research component/Fordwah |
1 |
|
20,000 |
|
20,000 |
|
Pakistan |
|
6. Environment Programme |
1 |
|
14,000 |
|
14,000 |
|
|
|
7. Health Programme |
1 |
51,000 |
|
|
51,000 |
DANIDA |
|
|
8. Africa Programme |
A |
159,000 |
|
|
159,000 |
AfDB |
|
|
9. Water and Environment in Asia |
3 |
65,000 |
|
|
65,000 |
IDRC |
|
|
10 WANA Programme |
4 |
76,000 |
|
|
76,000 |
IFAD |
Sudan/Pak |
|
TOTAL |
|
483,000 |
181,000 |
0 |
664,000 |
|
|
Country Projects
|
Projects |
Status |
Budget |
Budget |
Budget |
Total |
Donor |
Country |
|
Bangladesh: | |||||||
|
1. Research and Training on Irrigation Management |
1 |
|
|
83,000 |
83,000 |
FF |
Bangladesh |
|
2. Privatization of Minor Irrigation |
1 |
|
|
188,000 |
188,000 |
ADB |
Bangladesh |
|
India: | |||||||
|
3. M&E of Irrigation Management Turnover in India |
3 |
|
|
82,000 |
82,000 |
FF |
India |
|
4. National Water Management Project |
3 |
|
|
173,000 |
173,000 |
WB |
India |
|
Nepal: | |||||||
|
5. Water Rights and APP Project |
3 |
|
|
76,000 |
76,000 |
FF |
Nepal |
|
Sri Lanka: | |||||||
|
6. TA for Participatory Management of Irrigation Systems |
1 |
|
|
132,000 |
132,000 |
ADB |
Sri Lanka |
|
7. National Irrigation Rehabilitation Project |
1 |
|
|
282,000 |
282,000 |
EC |
Sri Lanka |
|
8. Shared Control of Natural Resources |
1 |
|
|
897,000 |
897,000 |
USAID |
Sri Lanka |
|
9. KOISP Evaluation |
1 |
|
|
94,000 |
94,000 |
GOSL |
Sri Lanka |
|
10 Irrigation Management and Crop Diversification |
1 |
|
|
22,000 |
22,000 |
ADB |
Sri Lanka |
|
11 TARC Project |
1 |
|
|
10,000 |
10,000 |
TARC |
Sri Lanka |
|
Burkina Faso: | |||||||
|
12 Improvement of Irrigation Performance |
1 |
|
|
160,000 |
160,000 |
AfDB |
B. Faso |
|
Niger: | |||||||
|
13 Irrigation Management Development in the Niger River Valley |
1 |
|
|
187,000 |
187,000 |
AfDB |
Niger |
|
Nigeria: | |||||||
|
14 Joint Management of Irrigation System - Phase II |
1 |
|
|
80,000 |
80,000 |
FF |
Nigeria |
|
15 Support to WUAs |
3 |
|
|
151,000 |
151,000 |
WB |
Nigeria |
|
Sudan: | |||||||
|
16 Supplementary Programme Support |
1 |
|
|
118,000 |
118,000 |
FF |
Sudan |
|
17 Latin America Programme |
1 |
|
|
152,000 |
152,000 |
FF |
Mexico |
|
TOTAL |
|
0 |
0 |
2,887,000 |
2,887,000 |
|
|
Pakistan Group
|
Project |
Status |
Budget |
Budget |
Budget |
Total |
Donor |
Country |
|
1. Direction |
1 |
|
109,000 |
|
109,000 |
|
|
|
2. Crop Based Irrigation |
1 |
|
|
131,400 |
131,400 |
ADB |
Pakistan |
|
3. Waterlogging & Salinity - Phase II |
3 |
|
|
132,000 |
132,000 |
Neth. Gov. |
Pakistan |
|
4. Pehur High Level Canal |
3 |
|
|
20,100 |
20,100 |
ADB |
Pakistan |
|
5. LBOD Project |
4 |
|
|
33,000 |
33,000 |
WB |
Pakistan |
|
6. Fordwah Eastern Saddqiya |
4 |
|
|
17,500 |
17,500 |
WB |
Pakistan |
|
7. Sailaba Irrigation |
4 |
|
|
93,000 |
93,000 |
IFAD |
Pakistan |
|
TOTAL |
|
0 |
109,000 |
427,000 |
536,000 |
|
|
> Status 3 grants are discounted by 33%
> Status 4 grants are discounted by 50%
Table 3.2 Research Staff at Headquarters and Field by Programme
|
Staff Category |
Programme |
Total | ||||
|
|
1 |
2 |
3 |
4 |
5 |
|
|
Headquarters | ||||||
|
International Staff |
3.5 |
0.0 |
1.0 |
4.0 |
3.5 |
12.0 |
|
National Professionals |
2.5 |
0.0 |
0.0 |
8.0 |
1.5 |
12.0 |
|
National General |
3.0 |
0.0 |
1.5 |
4.0 |
3.5 |
12.0 |
|
Field* | ||||||
|
International Staff |
5.1 |
0.0 |
3.7 |
4.1 |
5.1 |
18.0 |
|
National Professionals |
14.5 |
0.0 |
6.0 |
10.0 |
18.5 |
49.0 |
|
National General |
8.0 |
0.0 |
4.7 |
9.3 |
8.0 |
30.0 |
|
Total |
36.6 |
0.0 |
16.9 |
39.4 |
40.1 |
133.0 |
|
Legend: |
Programme 1: Assessing and improving the performance of irrigated agriculture |
|
|
Programme 2: Sector-level management of irrigated agriculture |
|
|
Programme 3: Improving public irrigation organization |
|
|
Programme 4: Toward local management of irrigation systems |
|
|
Programme 5: Sustainable management of water delivery and disposal |
* Country programme staff are not assigned by programme area, but normally work on a number of programme areas depending upon project requirements
In the past, most of IIMI's efforts were limited to reviews, comparisons, and evaluations of other peoples' work, and the conduct of performance case studies in a range of physical and institutional environments. Present and future activities aim at providing, in the next five years, a field-tested, comprehensive and workable set of performance indicators. This will be pursued on several sites through action in-depth work, including surveys of households and organizations. At the same time, these surveys will be used to test a variety of hypotheses on the determinants of performance such as the roles of tertiary development of canals and roads, rotational water delivery schemes, management information and decision support systems, and various forms of decentralization and user participation.
As research progresses with performance studies, understanding the linkages between water delivery, farming systems, farmers' responses, and agricultural output will become more important. Since behaviour mediates the relationship between system design, management, and policy changes on the one hand and changes in performance indicators on the other hand, focusing more explicitly on response of rural households, farmers and communities will be important.
(in 1994: zero research staff, zero core budget, and zero projects)
This programme focuses on the broader issues pertaining to sectoral policy toward irrigated agriculture. It included projects on the impact of alternative financing mechanisms on the performance of irrigated agriculture, priorities for public investment, and the development of methods for the participatory formulation of sectoral policy toward irrigated agriculture. This programme is relatively new and has only published 8 scholarly publications, 13 proceedings, and 2 country and working papers.
Research analyzed the performance of irrigation agencies across a large number of locations to identify the impact of payment of user fees on the performance of these organizations. It arrived at the important conclusion that, when these agencies are at least partially financially autonomous and dependent on user fees for their operating revenues, the payment of fees leads not only to greater availability of resources for O&M (operation and management), but more importantly to greater accountability to users and hence to sharply improved performance. This is an important research contribution which has circulated worldwide in well distributed publications and offers useful recommendations to policy makers and agency managers. The same question was analyzed with quantitative analytical techniques to show how the Philippine's National Irrigation Administration (NIA) responded to a reform that made it largely financially autonomous and dependent on user fees. It showed that this institutional reform induced NIA to be more cost efficient, to increase efficiency in water use, and to achieve greater equity.
Analysis of the pattern of public investment in irrigation in Sri Lanka showed that there has been a shift from investment in new projects to investment in rehabilitation and modernization, with emphasis on enhancing the efficiency of water management. Instead of charging water fees, Sri Lanka chose to delegate to users responsibility for O&M below main system level management. Investments in rehabilitation and modernization of water delivery systems, coupled with these reforms for participatory management, have had very high cost-benefit ratios, far higher than the previous investments in major infrastructure.
While reported by IIMI under this programme, the valuable research on the performance of public irrigation agencies logically pertains to Programme 3. Future research proposes to look at the bureaucratic and political processes through which irrigated agriculture policies are made.
Research in Programme 2 does not yet address the broader themes of sector-level management announced in the MTP, namely "to develop strategic planning methodologies and approaches for governments to formulate effective and realistic national and state policies and plans for the development and management of the irrigated agriculture sectors of their countries". We regard the research agenda of Programme 2 as of fundamental importance to IIMI in as much as it helps integrate irrigation management research with the broader sectoral and policy issues. Among those are the pressures to reallocate water to non-agricultural uses, which require increasing the efficiency of water use in agriculture and thus demand greater attention by governments to water management issues in agriculture. This new programme has yet to evolve a coherent research agenda, with well defined priorities and hypotheses, and a clear division of labour with other agencies conducting research on these broader themes. As soon as this is done, this programme needs to aggressively seek funding for its projects.
(in 1994:16.9 research staff, a core budget of US$ 185,000 and one core project)
This programme has published only 2 scholarly publications, 1 proceeding, and 1 working paper. However, most of the work reported under Programme 2 properly belongs here. In addition, a large share of country-programme research pertains to this area of work, particularly research done in Pakistan and Sri Lanka, even though there has been little formal integration with headquarters research programmes. Proposed research focuses on analyzing which kinds of public irrigation organizations work best, and how to manage the required processes of reform of these organizations. This would be done through comparative case studies using methods derived from the management and organizational literature. We regard this area of work as vital to IIMI's research agenda since public organizations at present dominate irrigated agriculture, and the reform of their management practices is a key to improved performance of irrigated agriculture. For success in this research, clear integration between country-programmes research and headquarters research programmes needs to be pursued vigorously.
(in 1994: 39.4 research staff, a core budget of US$ 907,000, and 13 core projects)
This is a subject in which there is considerable country and donor interest given the widespread on-going processes of transfer of irrigation management to WUA (Water Users Association) and the geographical importance of FMIS (Farmer-managed Irrigation Systems). It has published 6 scholarly publications, 27 proceedings, and 12 country and working papers, and the FMIS Newsletter in support of the FMIS network. This programme's budget includes much country-programme research which is not directly integrated with headquarters research.
The programme has addressed important questions on the determinants of effectiveness of local organizations for both surface and groundwater use, the role of supportive services and public assistance strategies to FMIS and the impact of public policy reforms such as water rights in affecting the performance of the FMIS sector, the correspondence between farmer management and the optimal technical design of water delivery systems, the development of approaches for the participatory generation of information and assessment of performance, and the design of local institutions for the management of village water tanks and underground water aquifers. The conditions for success in transferring irrigation management from government to local farmers' organizations is another important focus of research. It has been pursued through a number of country case studies (Philippines, Indonesia, Sri Lanka, Senegal, Sudan, Bangladesh, USA, and recently China) and subsequent comparative analysis and extraction of generic research results. This research has provided materials for the FMIS Newsletter and eight workshops of the FMIS network.
Important conclusions have been derived from this research, showing for example: (1) the importance of institutional, legal, and technical support for WUA, and hence the need for guidelines to define the relationship between public and private roles for effective water management; (2) the importance of extending the functions of WUA to the provision of other services in order to consolidate their financial viability and enhance their sustainability; (3) the mechanisms through which user participation in management of the irrigation agency that collects water fees enhances performance; (4) the design of alternative organizations to enlist the cooperation of and insure the coordination among users of underground water; (5) the development of specific procedures to gather the information necessary for the effective monitoring of user-managed irrigation systems; (6) the development of performance indicators specific to locally-managed irrigation systems; and (7) the development of methods for the self-assessment of WUA performance and the use of this information to improve performance. The forthcoming publication Locally Managed Irrigation Systems, combines IIMI's findings with the broader literature to extract generic management and policy implications.
In our judgment, this is an area of research and technical assistance that is one of IIMI's hallmarks and where it may establish itself in a leadership position within an international network of irrigation professionals and through the generation of a flow of new knowledge. The programme is pursuing an ambitious agenda of country case studies and comparative analyses covering several years of work. To give a truly new perspective that goes beyond a collection of facts, it will need strong intellectual leadership and greater coordination of efforts between headquarters and country levels. Research hypotheses and methodologies must be more strongly grounded in the important theoretical advances which have been made in the fields of institutional economics and organizational theory, in particular regarding the determinants of cooperation and the performance of local organizations, the design of agrarian institutions and contracts, and the determinants of household and organizational behaviour in the context of fragmented and incomplete markets. It also needs to result in a greater flow of professionally refereed publications. This area of research should be one of IIMI's strong points, given careful attention to the aspects described above.
(in 1994: 40.1 research staff, a core budget of US$ 664,000 and 10 projects)
Research efforts in this programme focus on the biophysical and engineering aspects, and its interface with the management aspects of irrigated agriculture. It has been very successful at linking these fields and should capitalize on this unusual integration. Studies are geared towards the understanding of how crop diversification, canal lining and maintenance, the use of decision-support tools, and the utilization of groundwater in conjunction with surface water affect the conservation of water, the flexibility and stability of water delivery, and the sustainability of the physical resource base. This programme has published 9 scholarly publications, 42 proceedings, and 15 country and working papers.
Through a number of collaborative efforts in Bangladesh, Indonesia, the Philippines, and Sri Lanka, IIMI studied the system management requirements and the incentive structure needed for crop diversification, in general away from rice toward more high value crops. The research showed that achieving diversification is a complex business that requires changes in the water delivery, economic, institutional, and social context within which farmers make their choices.
In Pakistan and Sri Lanka, IIMI studied the impact of interventions on water delivery in two ways. First, by studying in the field the effect of three interventions (lining, desilting, and changes in operational management) on water delivery performance, and second through the development of a decision-support tool for improving water delivery by irrigation canals. Some of the conclusions from this action research are: (1) canal lining is only effective if the lining process is of high standard or where groundwater is already saline; if not, selective desilting based on hydraulic modelling tends to be more cost effective; (2) systematic planning, implementation, and monitoring of maintenance is technically and economically feasible in a developing country context; (3) conditions for equity in water distribution can be improved when a systematic maintenance programme is followed and rehabilitation of distributary canals is done; and (4) supply and demand of water can be better balanced and water losses reduced by a combination of the following measures: restoration of canals, improvement of maintenance practices, use of decision-support systems, and in Sri Lanka at least the participation of WUA to O&M.
Preliminary research on decision-support tools has shown that there is considerable potential for these vastly underused techniques to help improve irrigation management. We suggest that IIMI should have a focused programme of scientific work in the field of information technology, and that this should be closely linked to the different research programmes. However, IIMI should not spend resources on creating specialty software, but continue to act as an intermediary between centres of excellence which develop software, and practitioners in the field.
In the Pakistan and India programmes, as well as in other country programmes, the conjunctive use of groundwater and canal water was examined. A number of key issues, such as aquifer drawdown, water quality, equity effects of water markets, and organizational problems have been addressed. This research showed how to improve the information base and the monitoring process for aquifer behaviour and environmental degradation, and how to craft institutions to manage groundwater irrigation at the level of aquifers or groups of tubewells in a larger water-management context.
In some of its country projects (Pakistan, Sri Lanka, and West Africa), IIMI staff has been studying the effects of irrigation management on waterlogging and secondary salinization which is the result of inadequate irrigation and/or irrigation with salt-containing water. While research has been of the diagnostic type, IIMI should now focus attention on methods and techniques through which salinity in the root zone can be maintained at a level compatible with the cropping system.
Whereas in the past most research activities in this programme were of the action-type and rather dispersed, these problems now need to be studied in a more quantitative and integrated fashion. Irrigation agencies are badly in need of methods and tools that combine advanced methods and monitoring techniques to improve the management of physical systems. A major contribution of IIMI in this field should consist of combining and adapting existing methods to make them operational in less developed countries. Integrated system models are very suitable instruments for predicting: (1) the effect of changes in O&M systems and management scenarios on water delivery in terms of efficiency, equity, adequacy, and reliability; (2) the cost-effectiveness of planned improvements in physical infrastructure; and (3) the downstream environmental impacts and the effects of upstream activities on the irrigation infrastructure. Optimization models make it possible to utilize and manage water resources for irrigation better. Simulation models can be used to analyze the processes controlling the physical functioning of an irrigation system and the relation between the system and the surrounding environment. We consequently suggest that IIMI should significantly strengthen its scientific skills in this area of research, and make more effective use of current collaborations to absorb this knowledge into its own research agenda.
3.3.6.1 Gender Issues
3.3.6.2 Health and Environmental Issues Related to Irrigation
3.3.6.3 Choice and Use of Technology
3.3.6.4 Improving the Global Database on Irrigated Agriculture
In agreement with the 1990 review, we stress the fundamental importance of gender roles in affecting the performance of irrigated agriculture and hence the need for IIMI to devote greater attention to this subject matter in its research. This theme is currently a subactivity within the Local Management Programme, and is staffed by an Associate Expert. Progress has been made in initiating research activities and conducting field work in Sri Lanka, Nepal, and Niger. However, it has as yet had little impact on IIMI's work and has resulted in little research output. In general, this programme suffers from insufficient commitment by permanent staff and insufficient interactions with the rest of IIMI's research, particularly at headquarters. There are no clear overall directions in IIMI's research toward gender and there is no integration with the overall research programme. This programme has suffered from IIMI's lack of attention to understanding behaviour, particularly at household and community levels. And not until the DR and programme leaders explore systematically how the gender dimension enters in their research will the theme be given due attention.
The Panel recommends that all programmes and projects be reviewed for research opportunities related to gender, and objectives be established based on these.
This recommendation will also require that serious consideration be given to recruitment of professional staff with the appropriate skills for this type of research.
Irrigation has been associated with several vector-borne diseases and the management of water flows can influence their incidence. IIMI has been cooperating with PEEM (Panel of Experts on Environmental Management of Vector Control) in defining a research agenda, but this is still at a very preliminary stage.
Work on the relation between irrigation and the environment focuses on impacts within the irrigation system (waterlogging and salinization), above and on the irrigation system (watershed management), and by the irrigation system (downstream pollution and disturbances to water flows). Clearly IIMI has done important work on the first type of impact and this it where it should maintain its main thrust. As it focuses more on irrigated agriculture, a number of additional within-system environmental issues should be addressed such as the relationship between water management, pest control, crop diversification and agrochemical distribution.
With the USAID-funded Shared Control of Natural Resources (SCOR) project, the Sri Lanka field programme is now establishing an action research programme that integrates watershed and irrigation processes and management. Due to the difficulties of measuring biophysical processes, work on upstream and downstream effects will have to be done cooperatively with other CGIAR centres and specialized research agencies. In addition, given the fundamental issue of creating economic incentives for upstream land users to manage their resources in less damaging fashions, strong expertise in resource economics is needed, and this would be a good basis for cooperation with IFPRI. We suggest that IIMI may find it more consistent with its current expertise in within-system management to move toward the downstream environmental effects as opposed to upstream issues in the first instance.
The functioning of an irrigation system is controlled by the physical characteristics of the water delivery system and by its management, where the latter is the result of interactions between farmers, water users associations, and the irrigation agency. For an irrigation system to perform better, it is thus necessary that both its physical and organizational components be simultaneously improved. For this reason, it is essential that IIMI staff maintain expertise in technological issues for work in all of the five main subject areas of its agenda. The physical interventions which have been studied are few, and limited to canal lining, desilting, and changes in maintenance procedures. Given the strong management-technology linkages, which should be a hallmark of IIMI's expertise in the CGIAR System, we suggest that IIMI should pay greater attention, in collaboration with others such as IPTRID, to new technological options in irrigation and drainage, to scientific decision-support methodologies, and to use of modem measurement and communication systems, and to continued developments in the modelling of transport processes within the soil.
Given its worldwide mandate, and particularly if it is to become a Systemwide Convening Centre on Water Management Research, IIMI should have a strong quantitative perspective on the global state of irrigated agriculture. A global irrigation database would assist IIMI in assessing emerging issues, and be useful to many other international organizations, less-developed countries researchers and decision-makers, and the private sector. Mainly due to lack of funding, no attention has been given to this cross-cutting theme. This is, however, an area where IIMI needs to move with caution, as there have been countless wasted efforts at building data bases. This is a major enterprise more in line with the FAO's capability and mandate. The Panel suggests that IIMI should not start collecting data before a study is done on the type of information that it would be relevant to collect, the method of data collection to be followed, and the data processing hardware that it would need to acquire. We do not regard it as an IIMI priority.
3.4.1 The Product of Adjustment to Strategic Research: Quantity and Quality of Research Output
3.4.2 The Process of Adjustment to Strategic Research
3.4.3 Needs in Adjusting to Strategic Research
A key question posed by TAC to this Review Panel concerns how far IIMI has gone in moving from a donor-driven project specific approach to a strategic demand-driven programme approach to research, after its entry to the CGIAR System, and how effectively this has been done. We answer this question by first characterizing the product expected from research in general and strategic research in particular. We then turn to the process used to achieve greater focus on strategic research. A linked question is how research has been affected by the broadening of IIMI's mandate from irrigation management to the management of irrigated agriculture.
In general, we find that IIMI has made modest and uneven progress in achieving strategic research since entry into the CGIAR system, and that overall research performance still lags behind potential in quantity, quality, and impact. A few research products have been excellent and have received considerable attention. They are, however, still too few and only weakly cumulative in the development of a broader IIMI view.
In quantity, refereed journal articles for the last five years numbered 5 in 1989, 6 in 1990, 8 in 1991, 4 in 1992, and 6 in 1993 (Table 3.3). The total number of scholarly publications (refereed journal articles, monographs, and research papers) numbered 7,6, 10, 4, and 9 in those same years, respectively. In-house publications numbered 19, 14, 6, 10, and 17, respectively. The totals of all publications were 38, 36, 32, 36, and 37, respectively. These figures show no significant change in research output since entry to the CGIAR or any greater focus on refereed journal articles and scholarly publications, where strategic research would be published. Although there is a time lag between adjusting priorities and achieving published output, there is no evidence that there is a larger flow of papers in the pipeline. Overall, this is a rather low output for a research staff of 43 professionals at headquarters and 91 in country programmes (Table 2.3), This is confirmed by a recent survey of publication rate in 22 IARCs, including 16 CGIAR Centres. IIMI ranks 21st for total publications per total research staff and 19th for journal articles per internationally recruited scientist per year. This low performance raises the important questions of whether there is proper time allocation between research and other activities, whether there are adequate incentives to stimulate publication, and whether there are the right professional skills to generate publishable research.
Table 3.3 IIMI's Publications by Outlets and Years
Classification 1:
|
Type of publication |
1989 |
1990 |
1991 |
1992 |
1993 |
Total |
|
Scholarly publications |
7 |
6 |
10 |
4 |
9 |
36 |
|
Proceedings |
17 |
20 |
19 |
26 |
17 |
99 |
|
Working & country papers |
14 |
10 |
3 |
6 |
11 |
44 |
|
Total |
38 |
36 |
32 |
36 |
37 |
179 |
Classification 2:
|
Type of publication |
1989 |
1990 |
1991 |
1992 |
1993 |
Total |
|
Refereed journals |
5 |
6 |
8 |
4 |
6 |
29 |
|
Chapters in proceedings elsewhere |
14 |
16 |
18 |
22 |
14 |
84 |
|
In-house publications |
19 |
14 |
6 |
10 |
17 |
66 |
|
Total |
38 |
36 |
32 |
36 |
37 |
179 |
|
Legend | |
|
Classification 1: |
Scholarly publications include journal articles, research papers, and monographs. Proceedings include chapters in conference proceedings and complete conference proceedings published by IIMI. |
|
Classification 2: |
Refereed journals are articles in international refereed journals. In-house publications include monographs, proceedings, and other publications published by IIMI. |
Quality of research is of course difficult to assess. Publishing refereed journal articles and scholarly materials is one test, particularly for strategic research, and it is poorly met. Another test is a survey of opinion of 47 national institutions run by the TAC Secretariat, a standard practice in CGIAR centre reviews. Results suggest that IIMI's research during the last five years is assessed to be between slightly and moderately valuable (see details in Appendix IV), an average score of 2.9 on a 5-point scale. These are modest marks at best. Scores given by research scientists and research administrators are lower than those given by managers and policy makers. Scores given by those who have been associated with IIMI for more than 3 years are, however, higher. Assessment of the future importance of IIMI's research areas is higher, with scores ranging between moderately and highly valuable, and an average of 3.8, confirming the importance of IIMI's research mandate and confidence among national institutions that it can deliver valuable research results in the future.
Our own assessment of research quality derives from our detailed review of a selection of IIMI's most important writings made for us by the Institute. In general, we find that individual papers address relevant issues. From the very first presentation of research results by programme heads to the review panel, we noted a tendency to undersell results. Undervaluation of achievements is compounded by scarcity of sufficiently visible publications. However, we also find that research procedures are in general not sufficiently explicitly grounded in the relevant bodies of theory, that insufficient high quality quantitative analysis is being used, and that results are not sufficiently translated into specific guidelines for management and policy improvements. This is partly due to the difficulty of the field of research, as explained in the first part of this chapter, but it mainly reflects an approach to research which is too often descriptive and discursive, lacks sufficient scientific rigour, and derives management and policy implications that are much too general. More use needs to be made of rigorous mathematical techniques, without implying that mathematics and statistics must permeate all research. More attention needs to be given to simulation of the impact of alternative physical, management and policy scenarios on performance indicators. For this, greater attention needs to be given to the specification of behaviour, to the identification of causal relations and processes, and to the use of analytical techniques that permit either comparison of alternative scenarios or derivation of optimum solutions. IIMI's research needs to be more solidly based on its vast cumulative experience, better grounded in theory and advanced methods of analysis, consistent with IIMI's mandate and with the institutional and informational context of the countries where it works.
Studies on irrigated agriculture are found in work on crop diversification, and on analyses of upstream (land use in watersheds) and downstream (effects of saline water in drainage) processes. We commend IIMI for not losing sight of its main focus of research while broadening its mandate from irrigation to irrigated agriculture. However, there are obviously many more important issues that concern irrigated agriculture than the above, and a comprehensive plan for work on this is still lacking. Some of these issues will enter in IIMI's research through a focus on farm household behaviour, farming systems, the interface between technological change in crops and water requirements, and work on sectoral policies.
Research has had an impact on the scientific community through several notable publications which have circulated widely and attracted the attention of the international scientific community to IIMI. This impact is however still modest. Impact on irrigation managers and policy makers has been obtained by raising their awareness of the large gap that exists between performance and potential, and by providing them with a number of alternative options to reduce this gap. By far the most significant impact has been obtained through the role of research in supporting institutional strengthening in specific country programmes and this is discussed in Chapter 4.
3.4.2.1 The Future Definition of Research Programmes
3.4.2.2 Achieving Better Performance in Research
The Panel suggests that current programmes continue to be in an excessive state of flux, and would benefit from some logical reordering and greater focusing. Since IIMI was started, the programme structure has been reorganized many times, with definition of some 30 successive alternative programmes over the years. Reorganization into a two-programme approach has recently been discussed as a way of focusing the research programme. We suggest that, with proper adjustments that can be accommodated within the current MTP, the programme structure in place seems generally adequate and should be given a chance to succeed instead of being amended once again so soon after definition. Better focusing of research should be achieved by sharper specification of priority research within each programme, rather than by administrative restructuring. Energies should now be fully devoted to implementing the current research agenda.
Five research programmes does not appear to be excessive as long as there are strong interactions across these programmes. If country-programmes personnel is affiliated to these programmes, critical mass is easily reached, although not necessarily in every country location. A 5-programme structure offers the advantage of stimulating decentralized initiatives by several research leaders. For donors, it offers clarity of the types of programmes which are being supported. And for research professionals, it offers identification with recognizable disciplinary traditions. However, as we shall discuss below and in Chapter 5, it only makes operational sense if it is accompanied by more decentralization of programme management compared to the current research management structure.
A suggested logical reordering of the current programme structure is as follows:
Programme 1: Water delivery, use. and disposal (currently Programme 5, "Sustainable Management of Water Delivery and Disposal"). This programme focuses on the interfaces between hydrology, engineering, agriculture and management; on the technology of water delivery, use, and disposal; on cost-benefit analyses of alternative options such as lining, desilting, drainage, and rehabilitation within different managerial structures; and on the development, adaptation, and use of quantitative decision-support tools. This programme should seize the opportunity of entering forcefully into environmental issues by capitalizing on a unique combination of interdisciplinary skills in biophysical and management sciences. Analysis should extend to downstream externalities and watershed management. If this theme is successfully developed, it may in due time spawn a separate programme on the environment and irrigated agriculture.
Programme 2: Resource management by farm households and communities (currently Programme 1, "Assessing and Improving the Performance of Irrigated Agriculture"). In accordance with the workplan set out in the MTP and the high priority given by the Board to this subject of inquiry, this program needs to complete the research in progress on performance indicators. This is expected to take up to 5 years more. After completion of this research agenda, further work on indicators should be present in all programmes, according to the issues addressed. In particular, analyses of the determinants of performance should be the objective of all programmes.
It is difficult to predict priorities so far ahead, especially in view of the likely changes in CGIAR. Below is sketched out an alternative agenda that could follow the current Programme 2. This would focus on behaviour in relation to management of resources, water in particular, since this is the linkage between changes in system design, public and private management of irrigation, and policy and changes in performance indicators. The relevant levels are: (1) Rural households and individuals within households by gender and age, stressing differentiation across household types. (2) Communities, particularly for the management of common property resources. This includes contractual arrangements among households and communities for the allocation of water. This Programme would be the most logical home for the gender theme in terms of disciplinary affinities.
Programme 3: Local organizations for the management of irrigation systems (currently Programme 4: "Toward the Local Management of Irrigation Systems"). This programme is currently properly defined. It focuses on intermediary organizations, principally water users associations (WUA), farmer-managed irrigation systems (FMIS), and cooperative organizations for the management of surface and underground water. It addresses the important issues of turnover and privatization of water management, and of complementary institutional reforms as governments cease to deliver services, including service cooperatives, credit unions, technical assistance services, etc.
Programme 4: Public irrigation agencies (currently Programme 3: "Improving Public Irrigation Organizations"). This programme focuses on administrative mechanisms of water allocation and on the determinants of performance of public irrigation organizations, for instance the roles of bureaucratic incentives, decision making hierarchies, and financial responsibility. The process (speed, sequencing, and complementary reforms) through which reforms are being implemented is also fundamental for the success of managerial reforms as well as for eventual privatization and transfer.
Because public and private sectors always coexist in the management of water, with different relative weights and functions, careful consideration should be given to bringing Programmes 3 and 4 into one single programme that deals simultaneously with public and private organizations for the allocation of water. Arguing in favour of keeping them separate, but closely interacting, is the fact that they rest on different intellectual traditions within the social sciences.
Programme 5: Sector-level management and policy toward irrigated agriculture (currently Programme 2: "Sector-Level Management of Irrigated Agriculture"). This programme would focus on issues of public finance and investment priorities, planning techniques for the intersectoral allocation of water, property rights and the legal foundations of water contracts, the development of water markets and water pricing as mechanisms for the allocation of water, regulatory agencies for the performance of water markets and the determination of water fees, regulation of pollution and incentives for sustainability, the broader policy framework determining the structure of incentives and constraints, and long run dynamic forces such as population growth and global environmental change. Much of this work should be done cooperatively with IFPRI, the World Bank, and other agencies (see Chapter 7). It is, however, important for IIMI to participate at this level of analysis in order to maintain a link with the broader picture of water management.
There currently exist large imbalances across programmes in terms of staffing, resource support, and level of activity. We suggest that, if IIMI has a serious commitment to its current MTP, careful attention must be given to the discrepancies between desired and actual programme sizes and that remedial actions be correspondingly taken.
Cross-cutting themes were recently introduced, but have as yet not been a success. This is typical of a matrix approach to research management, where the secondary dimension (themes) is either competing for resources with the primary dimension (programmes), or left unstaffed and neglected. In our opinion, cross-cutting themes have not been organized in such a way that they can work. We suggest that these themes should be vigorously pursued but that this should be done by integrating them in all programmes when relevant, as opposed to de-facto managing them as separate sub-programmes. Each programme leader should have clearly formulated plans as to what his/her programme should be doing about cross-cutting themes (see Chapter 5).
IIMI should give far greater attention to the link between irrigated agriculture and the environment as this theme emerges as a well acknowledged critical issue and the CGIAR has identified it as one central to its mandate. Failure to move pro-actively on this theme exposes IIMI to the risk of finding itself dangerously lagging at a later date. Greater attention should be given to finding opportunities to do research on the environment within IIMI's mandate and core budget constraint, and collaboration with other institutions working on this vast subject should be arranged. IIMI should prepare itself for its role as a 'Systemwide convening Centre on water management research' by being proactive in addressing related environmental issues. A separate programme on the environment and irrigated agriculture may well be organized at a later date if this theme is successfully developed.
The Panel recommends that IIMI explore opportunities to expand its research on the environment, particularly in relation to the Systemwide initiative on water management.
This Panel is conscious of the danger of suggesting additional themes in a context of, at best, zero-sum core budgets. Trade-offs will consequently need to be made with other existing programmes. To this end, IIMI will have to engage in a full review of its priorities. Possible sources of savings are the consolidation of Programmes 3 (Public Organizations) and 4 (Local Management), minimal expenditures on Programme 2 (Sector) by conducting research principally through cooperation with others, and extensive collaborative agreements in the environmental field as well. In our judgement, however, the main sources of funds for this additional area of work should come from enhanced internal productivity in research, following the recommendations made by this Panel, which ought to allow work in new areas without sacrificing output in the existing programmes. Finally, if IIMI makes a convincing demonstration of its specific capabilities in this important field, additional funding through contract research seems to be a realistic possibility.
This Panel is also conscious of the need for IIMI to prioritize its efforts between moving into new research themes and enhancing the quantity and quality of its strategic research as recommended by this Panel. We feel, however, that progress in the environmental theme is too rapid and widespread for IIMI to lag behind without serious risks.
IIMI's research projects need to relate more explicitly to the broader global picture, to identification of the major determinants of a water "scarcity" crisis and of the sub-optimal performance of irrigated agriculture. The relatively poor definition of current research is not due to lack of vision, but to the width of the gap between global vision and specific project definition. Establishing relationships between the broader picture and research priorities is the fundamental purpose of the development of a long term strategic plan and of an MTP which makes a collective vision operational. To be effective, future MTPs consequently need to be developed through intensive participation and cooperation among research programme leaders, country programme heads, and management and their respective teams. Only if the MTP is the outcome of such a process can it become a consensus document that sets research priorities, coordinates research, gives guidelines for project definition and screening, and relates project initiatives to the broader picture.
Given the rather unique approach which IIMI has chosen to fulfill its mandate, achieving greater performance in research also requires that headquarters research and country programmes be integrated into a single coherent programme, which has research, institution strengthening, and training components. Definition of research priorities should in part be done in response to identification of critical bottlenecks and opportunities in irrigation management at the level of country programmes. Information gathering and applied research done in country programmes should be clearly linked with overall research objectives and with Institute-wide research programmes. IIMI's greatest advantage in insuring relevance to its research is the close link which it has with user-driven demands. However: (1) Close participation of national representatives in the definition of country programmes through local Consultative Committees (CC), while essential, should not be confused with the setting of research priorities toward strategic research. CC demands for information may tend to be dominated by the need to provide answers to short run crises. (2) Headquarters needs to provide greater scientific guidance and more precise definition of research methods, in order to permit effective comparative analysis across projects and countries, and the extraction of strategic knowledge from country-programme action research. (3) The current separation between research and institution strengthening in defining the content of country programmes is reinforced by the separate control over these two functions by a Director of Research and a Director of International Cooperation (DIC).
Most country-level institutional strengthening activities that have a sufficiently specific focus should be explicitly identified with IIMI's research programmes, either/or both as sources of information or for the dissemination of research results. In addition, all country research activities should be part of an IIMI research programme and hence under the shared authority of the corresponding research programme leader and the country programme head. A suggested organizational structure and a set of operational procedures in support of this integration are presented in Chapter 5. In this chart, the DDG assumes the responsibilities formerly assumed by the DIC and some of those by the director for Pakistan.
The Panel recommends that each country research activity and most institutional strengthening activities be associated with a research programme, with the Director of Research and the Deputy Director General coordinating this integration.
As CGIAR financial contributions have declined, and IIMI, like other CGIAR Centres, has to rely increasingly on contract project funds, the risk of "following the path of a consulting firm" is enhanced, particularly since opportunities for contracts are relatively abundant in this area of expertise. There is in principle no reason why reliance on complementary funding cannot be made consistent with IIMI's mandate, implementation of a coherent research programme, and achievement of strategic research. However, this will require increasingly careful adherence to sharply defined programme priorities set in the MTP, clear definition of hypotheses and methods before acceptance of special projects, and continuity of effort and accumulation of knowledge. To achieve this necessary guidance, we make a recommendation in Chapter 5 on the Programme and Project Subcommittee.
We noted that the volume of publications must be increased and that more research results need to be published in professional journals and refereed outlets. At the moment, too many research results remain in the grey literature where they are difficult to access and are not subjected to the rigors of anonymous peer review. For this purpose, the proper incentive structure must be put in place. In general, there should be much greater explicit demands made of the research staff to publish, with the highest scientific standards relevant to fulfill IIMI's mandate. In discussions with several top IIMI scientists, we noted confusion between the need to publish in the scientific literature, and perceived relevance of this type of knowledge for real practitioners in the field to whom IIMI scientists have strong commitment. This confusion needs to be resolved, and a clear definition of the requirements for publication should be specified in job descriptions and in the definition of goals. Evaluations of research staff performance should be done, with feedback to staff directed at helping individuals improve their future research performance. Associate experts seconded to IIMI should be explicitly induced to publish on behalf of the Institute and generally jointly with their IIMI collaborators. For the moment, those experts located at headquarters are not sufficiently integrated in IIMI's research programmes and are too often left adrift. This is not the case in country programmes where this integration has usually been effective. Greater incentives to publish should also be given to national collaborators by defining more explicitly their roles in the process of research and publication. In general, these collaborators should be joint authors of the publications based on work in which they have participated.
As recommended by the two previous review panels, greater use should be made of external panels of experts and international advisors to enhance quality of research output. According to IIMI's response to the 1990 donor sponsored review (see Appendix VI), this does not appear to have been done except on one occasion. It is, nevertheless, a small part of what is needed to insure quality. The key sources of quality performance and quality control are in our opinion to be sought in improving the intellectual climate that prevails at IIMI. This climate must stimulate creativity and collegiality, and use peer pressure to enhance the quantity of research and peer advice to enhance the quality of research. It should be created by organizing periodic seminars, informal presentations of proposed research initiatives and early results, reading groups on new subjects to be explored, short courses on key issues (e.g., new advances in quantitative analysis and computer packages, in the theory of institutions, in management techniques, etc.), and in general by fomenting an atmosphere of intellectual ferment and discussion. At present, not much of this occurs, and individualism is the dominant norm.
Sharper focusing of research also needs more clarity in the division of labour with other research institutions in the field of irrigated agriculture and the necessary supporting collaborative arrangements. These arrangements have to be based on a clear understanding that IIMI's five research programmes all need specific partners. This includes both other CGIAR Centres and other participants to the field (see Chapter 7).
Finally, quality can be increased by a stronger interface with the international scientific community and with centres of excellence at the frontier of knowledge in the different fields involved. For this, IIMI has to seek greater involvement in collaborative projects, and make more effective use of visiting scholars, students doing doctoral dissertation research, and scientists on sabbatical leave. Participation in scientific conferences should also be encouraged.
Strategic research is a demanding task that requires technical skills and the organization of research teams that are markedly different from those needed for donor-driven project-specific research. We started this chapter by observing that the subject of irrigation management research is highly interdisciplinary. This raises difficult questions for the disciplinary composition of IIMI's research team, as it does in other CGIAR Centres which focus on agroecological systems, research management, and policy analysis. Since scientists need to interact with peers to keep up their disciplinary skills, this argues in favour of limiting the disciplines represented at IIMI to the essential ones. Disciplinary upkeep is partially insured by membership in at least one of the five research programmes, which each have disciplinary traditions. Access to other disciplinary talents needed to conduct interdisciplinary research should be sought through cooperative arrangements with top experts outside IIMI rather than by attempting to achieve full intellectual self-sufficiency. However, at least one IIMI staff member must be sufficiently qualified in the discipline to understand the scientific knowledge sought externally. As IIMI is to become a convening centre for a 'Systemwide initiative on water management' research, this disciplinary composition may have to be cautiously broadened.
Reading IIMI's research suggests that the quality of economic analysis is in general inadequate. Performance indicators include a variety of economic criteria, but a rigorous conceptualization of the processes that change performance levels is lacking. The skills needed to conduct rigorous social science analysis, economics in particular, are not possessed by enough of the current research staff. This raises important issues for upgrading the level of skills of current staff and selecting priorities for future recruitment of professional staff for the strategic research that is planned.
Ultimately, the Institute's only guarantee of improved strategic and overall research performance rests in the quality of its research personnel, and especially of its DR and research programme leaders. These leaders have to be given greater incentives to perform and hence greater ownership over their programmes. To this end, we suggest that the decision-making processes through which research projects are defined, external financial support for contract projects is sought, research teams are organized and staffed, and the relevance and quality of research is monitored should be more decentralized to the level of programme leaders. This greater degree of autonomy and responsibility of programme leaders is also needed for IIMI to achieve more success in generating desirable research contracts in an increasingly competitive arena of grants and special projects (Chapter 5). However, we stress that the suggested reorganization requires that programme leaders be first class international research scientists, research administrators and motivators of people, an unusual and demanding combination of talents. IIMI's success at strategic research will depend fundamentally on the quality of their performance, because decentralization has to be earned. Superior research and management performance will thus be expected from these leaders, after the opportunity to perform in a more decentralized fashion has been offered to them. We are optimistic that this can indeed be successfully achieved if these steps are taken.
