Research Report

International Water Management Institute

INTERNATIONAL WATER MANAGEMENT INSTITUTEP O Box 2075, Colombo, Sri Lanka

Tel (94-1) 867404 • Fax (94-1) 866854 • E-mail iwmi@cgiar.orgInternet Home Page http: //www.cgiar.org/iwmi

ISSN 1026-0862

36

ISBN 92-9090-387-2

Institutional Change and SharedManagement of Water Resourcesin Large Canal Systems: Resultsof an Action Research Program inPakistan

D. J. Bandaragoda

Research Reports

IWMI’s mission is to contribute to food security and poverty eradication by fosteringsustainable increases in the productivity of water through better management of irri-gation and other water uses in river basins. In serving this mission, IWMI concentrateson the integration of policies, technologies and management systems to achieve work-able solutions to real problems—practical, relevant results in the field of irrigation andwater resources.

The publications in this series cover a wide range of subjects—from computermodeling to experience with water user associations—and vary in content from directlyapplicable research to more basic studies, on which applied work ultimately depends.Some research reports are narrowly focused, analytical, and detailed empirical stud-ies; others are wide-ranging and synthetic overviews of generic problems.

Although most of the reports are published by IWMI staff and their collaborators,we welcome contributions from others. Each report is reviewed internally by IWMI’sown staff and Fellows, and by external reviewers. The reports are published and dis-tributed both in hard copy and electronically (http://www.cgiar.org/iwmi) and wherepossible all data and analyses will be available as separate downloadable files. Re-ports may be copied freely and cited with due acknowledgment.

ii

Research Report 36

Institutional Change and SharedManagement of Water Resources in LargeCanal Systems: Results of an ActionResearch Program in Pakistan

D. J. Bandaragoda

International Water Management InstituteP O Box 2075, Colombo, Sri Lanka

ii

Bandaragoda, D. J. 1999. Institutional change and shared management of waterresources in large canal systems: Results of an action research program in Pakistan.Research Report 36. Colombo, Sri Lanka: International Water Management Institute.

/ water resources management / irrigation canals / property rights / social organization/ social participation / leadership / landownership / farmer associations / water userassociations / equity / farmer-agency interactions / Pakistan /

ISBN 92-9090-387-2ISSN 1026-0862

© IWMI, 1999. All rights reserved.

Responsibility for the contents of this publication rests with the author.

The International Irrigation Management Institute, one of sixteen centers supported bythe Consultative Group on International Agricultural Research (CGIAR), was incorpo-rated by an Act of Parliament in Sri Lanka. The Act is currently under amendment toread as International Water Management Institute (IWMI).

The author: D. J. Bandaragoda is Senior Management Specialist of the InternationalWater Management Institute.

IWMI gratefully acknowledges the financial support for this study from the Govern-ment of the Royal Netherlands (Managing Irrigation for Environmentally SustainableAgriculture in Pakistan) and the World Bank (Pilot Project for Farmer-ManagedIrrigated Agriculture under the Left Bank Outfall Drain Stage I Project, Pakistan).Core funding to IWMI is provided by the Governments of Australia, Canada, China,Denmark, France, Germany, the Netherlands, and the United States of America;and by the Ford Foundation and the World Bank.

iiiiii

Contents

Acknowledgements iv

Summary v

Introduction 1

Objectives and Assumptions 2

Concepts Underlying a New Approach 4

Contextually Appropriate Methodologies 7

Results and Discussion 11

Conclusion 24

Appendix 26

Literature Cited 31

iv

The action research would not have beenpossible without the support of the staff ofirrigation-related agencies in the two provinces,Punjab and Sindh, where the pilot sites werelocated. There was a great deal of understanding,which developed gradually, between the agencystaff and study teams, and the author thanks allof them for being able to finish three of the fourphases of action research as planned. Specialthanks go to Professor Gaylord Skogerboe,Mehmoodul Hassan, Yameen Memon, Zafar IqbalMirza, Waheed uz Zaman, and field teams

Acknowledgements

deployed at the four pilot sites for their valuablecontributions in the action research program.

The author is indebted to the socialorganization volunteers, organizational leaders,and all other water users, who participated in thisaction research directly and indirectly, and tooksome risk in terms of pressure from personsresisting any form of institutional change. Finally,the author wishes to thank Douglas J. Merrey,Tushaar Shah, and Bryan Bruns for their valuableeditorial comments and suggestions on the earlierdraft versions of this report.

v

Summary

An action research program conducted at fourpilot sites in Pakistan found that organizing waterusers at the secondary level of Pakistan’scontiguous canal irrigation system was sociallyfeasible. This was contrary to the popular beliefsthat existed both within and outside Pakistan. Thepopular notions, which were related to constraintsof an integrated socio-technical system, illiteratefarmers, social pressure from big landowners andobstacles caused by the hierarchical society, wereproven to be invalid under conditions of aparticipatory process of social organization. Themethodology used was characterized by a step-wise process of social organization, which wascatalyzed by a locally recruited small field teamwith the assistance of community-based socialorganization volunteers. Training and other formsof capacity building were the major motivatinginfluences. A field implementation coordinationcommittee consisting of representatives of allservice delivery agencies working in the area,along with selected water users, highlighted thenecessity of farmer-agency coordination, andgreatly facilitated an incentive mechanism throughcollaborative activities. This combined effortresulted in successfully achieving the formation oftwo hundred water user associations (WUAs) atthe tertiary (watercourse) level, and four wateruser federations (WUFs) at the secondary canal(distributary) level at the four pilot sites.

The suggested institutional framework forsharing responsibility for water resourcesmanagement in large canal systems is acombination of property rights regimes. Notably,the newly introduced element to this framework isthe conversion of the existing state-propertyregime at the secondary canal level to acommon-property regime. This change will involvethe common-property regime at the secondarycanal level to have interactions with the state for

water delivery in the main canal and with aprivate-property regime at the tertiary level forappropriation of water resources units.

The emerging results of this social experimentare encouraging. The new WUFs were able to takecollective decisions to negotiate with state irrigationauthorities on joint management agreements formanaging water resources in the canal system.Although these agreements were not made imme-diately effective due to a procedural difficultyimposed by the present legal framework, the WUFsproceeded to test their capacity in undertaking aplanned maintenance program during the canalclosure period, and also to initiate a maintenance-related infrastructure improvement program.Replicability of this social organization program isenhanced by the methodology adopted: thedeployment of small field teams and the use oflocal volunteers. The step-wise social organizationprocess enhances sustainability. One drawback,however, has been the lack of full commitmentfrom the related government agencies, which haveto take the initiative for empowering the water userorganizations. Both the enthusiasm and thecapacity demonstrated by the water users in socialorganization for collective action show a goodpotential for further progress. Very likely, thedemand generated at the local level could facilitatea process of bureaucratic reorientation, which inturn would provide the necessary institutionalsupport for the new water user organizations(WUOs).

This report presents the conceptual andmethodological framework underlying the actionresearch effort. It has been validated in thecontext of a social environment that ischaracterized by a hierarchical society with a lowliteracy rate, but with a tremendous enthusiasm toabsorb new technology and gain higherproductivity.

1

Introduction

This report is based on the results of an actionresearch program in Pakistan conducted by theInternational Water Management Institute (IWMI).An analysis of empirical data is presented in thelight of existing theoretical knowledge on collectiveaction for natural resources management. Thereport also highlights the current constraintsassociated with a wider application of some ofthese findings, which reflect the difficulties inpursuing large-scale institutional reforms in therural sector of developing countries. Finally, thereport raises some key research issues that needto be explored further.

The action research program was based onthree main propositions, which were deviationsfrom conventional wisdom. First, it assumed that“farmers can manage,” which is a markeddeviation from the popular notion, particularlyamong government officials, that the farmers inthis region are mostly illiterate and incapable ofundertaking management responsibilities foroperating large canal irrigation systems (Haq andShahid 1997). Second, in contrast to the currentinstitutional strategies that normally emphasize arelatively narrow objective of reducing governmentcosts in managing irrigation infrastructure, theaction research program aimed at broaderresource management goals. Third, the actionresearch also sought to identify a demand-drivenbottom-up approach in establishing mechanismsfor decentralized management of water resources.This strategy was expected to address the issue

Institutional Change and Shared Management ofWater Resources in Large Canal Systems: Results ofan Action Research Program in Pakistan

D. J. Bandaragoda

of increasing disappointment with the impact ofthe usual top-down approaches for organizingwater users and transferring managementresponsibilities (Zaman and Bandaragoda 1996;Vermillion and Garcés-Restrepo 1998; Kikuchi,Fujita, and Hayami 1999; Samad and Vermillion1999; and Samad and Smidt 1999).

The main objective of this action researchwas to test, through pilot efforts, the viability offarmers organizations for managing parts of thewater resources systems so that more efficientand equitable use of water can be achieved in ahierarchical society such as in Pakistan. Toachieve this objective, the action researchconducted a number of field activities in aparticipatory manner. In the given context of large-size canal systems, and the deep-rooted socialperceptions regarding the role of the state as abenefactor and that of the water users as thebeneficiaries, the strategy of working towardsshared management was found to be veryproductive. To both the state agencies and thewater users, the idea of a complete managementtransfer to the user organizations at this stagewas not readily acceptable. The action researchprogram concluded that a property-rights basis forpossible joint management arrangements is aconceptually sound strategy to pursue in thesecircumstances.

This action research program coincided with apolicy resolve in Pakistan and several othercountries in the region to introduce major reformsaimed at improving the effectiveness of waterresources management institutions. The report

2

focuses on the process of organizing water usersin the context of responses of the main actorsinvolved in water resources management tochanges being proposed and introduced inPakistan. The contextual background in which the

action research has been designed is summarizedin the appendix. The report presents the newmethods adopted in this action research, and adiscussion on some emerging results.

Objectives and Assumptions

Action Research Design

The action research program, which was designedon the basis of the historical and contextualbackground outlined in the appendix, sought toanswer the following research questions:

1. Can effective WUOs be established in thegiven socioeconomic and technical contexts?

2. Can the capacity of the WUO members in thepilot sites be improved for taking collectivechoice decisions and actions related toimproved water resources management forirrigated agriculture? and

3. Can these WUOs and state irrigation-relatedagencies agree on their roles and functions ina situation in which WUOs decide toundertake secondary canal-level waterresources management on the basis of jointmanagement agreements between irrigationauthorities and themselves?

The broad conceptual approach underlying theaction research included some assumptions in theform of specific expectations for the futureprogress of participatory management inPakistan’s irrigation canal system. Theseexpectations were:

� that the WUOs would have a well-definedright to the water and authority over itsuse, and eventually be accountable for the

water received at the head of distributarycanals

� that the WUOs would be responsible fordistribution of water among the member wateruser associations (WUAs) at the watercourselevel according to their own agreed allocationrules

� that the WUOs and their member WUAs wouldbe ready to be responsible for managinggroundwater levels in their respectivecommand areas

� that the WUOs would reach an agreement withtheir members, as well as with the agencies,for assessment and collection of appropriatewater charges and/or operation andmaintenance (O&M) costs of irrigation anddrainage facilities in their distributarycommand areas

� that they would undertake the collection ofwater/drainage charges, improve watermanagement practices, and carry out otheractivities related to the use and disposal ofirrigation water, including the maintenance ofirrigation and drainage facilities

� that the government would introduce proposedinstitutional changes in the form ofsemiautonomous Provincial Irrigation andDrainage Authorities (PIDAs) in the fourprovinces, and Area Water Boards (AWBs) at

3

the main canal level, at least on a pilot-scaleinitially

� that these PIDAs and AWBs would providethe necessary institutional support to thenewly established WUOs, and together, theywould form an improved institutionalframework for irrigated agriculture in Pakistan

To realize these expectations, the projectdesign further anticipated that the WUOs would beable to develop and enforce appropriate internalbylaws, which would be binding on their members,and resolve any water-related disputes that mayarise among them. It was envisaged that some“social engineering” by the social organizers wouldbe able to catalyze this process so that theWUOs and their members would agree upon a setof rules, rights, and responsibilities.

During initial interactions with agency staff andfarmers, several other assumptions were alsomade. The main items were:

� that the operating agencies would be ready toempower the new pilot WUOs and cooperatewith them to ensure uninterrupted watersupplies and O&M of the physical systems(the requirement for government agencies’commitment in this regard was not reflected inany of the project documents)

� that the government would assist the newWUOs to enforce their internal rules byproviding them with an adequate enabling legalenvironment

� that the individual water users would derivesome economic gain out of being organizedfor taking over additional responsibility(farmers often question the potential ofindividual economic gain from collectiveaction)

� that the organized farmers could cope with theexisting social pressure and political and

feudalistic forces, and collectively act toimprove equity in water distribution

Selection of Pilot Sites

In many action research efforts, a contentiousissue is the representativeness of the selectedpilot sites. Given the project-related constraints onchoice, an assumption was made thatconsideration of a few important criteria, such asdesign discharge, canal length, number of outletsand command area, distribution of water users interms of landholding size, and location of thesecondary canal along the main canal, would leadto a reasonably representative choice of pilotsites.

As this action research program was based ondonor initiatives in two major infrastructuredevelopment projects, the study sites had to benecessarily located in those two project areas.In the Punjab Province, the study had to belimited to one site within the World Bank-fundedFordwah Eastern Sadiqia (South) Irrigation andDrainage Project area, and in consultation withirrigation-related agencies, the Hakra 4-RDistributary was selected as the pilot site. In theSindh Province, three sites were selected from theLeft Bank Outfall Drain (LBOD) project area, onefrom each of the three “LBOD districts:” BarejiDistributary in Mirpukhas, Dhoro Naro Minor inNawabshah, and Heran Distributary in Sanghar. Inthis instance, the process of site selection wasdesigned during the study inception to ensure afairly representative selection, given the restrictionof the number of sites to three, and accordingly,the sites were selected in a participatory manner,involving both the operating agencies and thewater users. Locations of the four distributarycanal sites are shown in figure 1. The mainphysical and socioeconomic characteristics ofPakistan’s irrigation outlined in the appendix arecommonly applicable to all of these selected pilotsites.

4

FIGURE 1.Location map of study sites.

emphasize two main management responsibilities:system operation and system maintenance.Guided by such emphasis, developing-countryirrigation agencies are mostly concerned withoperating and maintaining the “irrigation schemes,”rather than with managing water as a resource.The rare occasions of conducting performancestudies using performance indicators are alsolimited to evaluating the “behavior” of physicalsystems. Recently introduced concepts and

Broader Outlook

The action research was based on the propositionthat a broader focus on water resourcesmanagement for irrigated agriculture, rather thanthe usual narrow focus on irrigation systemmanagement, would prove appropriate foreffectively organizing water users.

Many donor-related projects and governmentinterventions for investment in irrigation tend to

Concepts Underlying a New Approach

5

strategies of “participatory irrigation management”also retain this narrower focus on systemmanagement. Popular participation is oftenbelieved to be a strategy primarily aimed atincreasing the probability of establishinginfrastructure that people want, in ways peoplecan and will manage them (Meinzen-Dick,Reindinger, and Manzardo 1995).

In this perception, the major benefits fromusers’ participation through collective action arepresented as increasing the potential for bettermaintenance of physical facilities and reducing thefinancial and management burdens of the govern-ments. A number of countries adopted this proposi-tion (Turral 1995). Pakistan’s experience (seeappendix) shows that developing-country govern-ments usually tend to be content with satisfyingdonor pressure on these preferred emphases,whereas the donors also perceive the countries’fiscal and management constraints as their majormotivating factors for change. A common percep-tion is that “governments are finding that byinvolving strong water users associations in projectmanagement and fee collection at the local level,they can use the capacity of community membersto exert social pressure on their neighbors to pay”(World Bank 1993:57).

Initial interactions with farmers indicated thatthese apparently extractive and restrictiveobjectives would not encourage water users towork towards social organization for participatorymanagement. The need to improve fee collectionamong them, or the need to reduce governmentexpenditures, would not readily motivate individualwater users to participate in collective actions orjoint organizations. In contrast, the broader notionof improved management of water resourcesavailable for agriculture was more readilyacceptable to the farmers who were looking forincreased productivity in agricultural pursuits.

The broader objective of managing waterresources for agriculture has a wider connotation,which transcends the boundaries of “irrigationsystem management.” First, it encourages thinkingabout water as a resource, which is distributedamong, or appropriated by, a competing group ofusers within integrated water resource systems(Keller, Keller, and Seckler 1996). Second, itcompels the individuals to be concerned about thesustainability of water resource use in the longterm. Third, it alludes to a socioeconomic impactassociated with the use of water as an importantinput for a production process. Also, the wateruser is encouraged to think of water saving,water disposal, and water-related environmentalissues.

A Property-Rights Perspective

The attempt to broaden the traditional focus oforganizing water users and transcend beyond thephysical system was compatible with anotherconcept included in the action research, theproperty-rights perspective of shared managementin large canal systems. Whenever a waterresources system is jointly owned, it can betreated as a common pool resource,1 out of whichan individual consumer or appropriator wouldobtain resource units in such a way that eachbundle of resource units consumed is subtractedfrom the pool of resources, and is not available toother consumers. Scheumann (1997:34) illustrateshow common pool resource systems can be heldunder four different property-rights regimes, andthe circumstances under which overuse can occur,and when it can be avoided. On the basis of thisexplanation, water resources systems can also beheld by the following four categories of propertyrights:

1A concise definition of the term “common pool resources” (CPRs) given by Ostrom (1992) is helpful to understand the dynamics of aproperty-rights regime. CPRs are “natural and man-made resources sufficiently large that it is costly to exclude users from obtainingsubtractable resource units.” A distinction is also made between the flow of resource units and the resource system producing the flow.While subtractability is a characteristic of the resource unit appropriated from a CPR, the jointness of use is a characteristic of theresource system.

6

Private Property (e.g., private tube well, warabandi2

turn in a watercourse)

� Exclusion of consumers can be easilyapplied.

� Resource allocation can be efficientlyeffected.

State Property (e.g., main, branch, and distributarycanal water)

� Relatively high cost of exclusion can leadto free riding.

� Noncooperation of consumers is common.� Susceptible to overuse.

Common Property (e.g., village pond, farmer-man-aged irrigation system)

� Exclusion is possible.� Cooperation by consumers can be

achieved.� Resource allocation can be made efficient.

Open Access (e.g., unsupervised lake or largecanal)

� Non-exclusion is the norm.� Cooperation by consumers is not needed,

or difficult to achieve.� Overuse is most likely to occur.

The main features indicated for each categoryare specifically applicable to property rights relatedto water resources systems. Of these four formsof property rights, common property appears to bethe most applicable form to a common pool ofwater in a canal system, from which resourceunits are extracted by individuals on the basis ofan agreed system of rules. Common property

refers to situations where there is “tacitcooperation by individual users according to acomplex set of rules specifying rights of joint use”(Runge 1992:18).

On this basis, a canal irrigation system, inwhich a time-based water turn rotation (warabandi)among the individual water users is in operation atthe tertiary level, can be analyzed in terms of theabove categorization:

� The tertiary level watercourse is a combinationof common and private property rights. Thephysical system is common property as itbelongs to the whole group of water users inthe watercourse command area, whereas, thewater flowing at a given time in thewatercourse is private property as it belongsto one water user who is having the warabanditurn at the time.

� The secondary level distributary canal and theprimary level main canal, including both thewater flowing in them and their physicalfacilities, are state property.

Some would assert that common propertyarrangements in developing-countries failed dueto overuse or misuse, and eventual degradationof resources. As a remedy, private propertyregimes were imposed to replace commonproperty arrangements. However, these changesalso failed to stop the overuse of resources andeven resulted in “increased inequity in alreadyunequal distribution of wealth” (Runge 1992). InPakistan, both state and private property regimesover water resources have clearly failed inefficient and equitable water resourcesmanagement. The former case can be seen inthe way distributary canal subsystems aremanaged allowing for rampant “free riding” (Mirza

2Warabandi is a time-based rotational method, which is designed to achieve equitable distribution of water available for a watercourse.The rotation is by water turns fixed according to a predetermined schedule specifying the day, time, and duration of supply to eachirrigator in proportion to the size of the irrigator’s landholding (Bandaragoda and Rehman 1995). For each watercourse, there is awarabandi list giving the names of actual water users taking water from the sanctioned farm outlets along the watercourse, and thecorresponding time turns allocated to each water user.

7

1989:15), whereas, the latter is well illustrated bythe overuse of groundwater resources throughprivate tube wells (Meinzen-Dick 1996). Thehypothesis is that a well-coordinated effort

background; 2) use of community-basedvolunteers; 3) nondependence on externally fundedphysical improvements as an incentive; and 4) astep-wise social organization process.

Locally Recruited Small Field Teams asCatalysts

The preliminary stages of the action researchprogram found that a community that is normallysuspicious about outsiders, strangers, and newideas, preferred to listen to local opinion leaders.The project staff who were directly involved infield activities were all locally recruited, and wereable to break this barrier of mistrust.

Two other characteristics of the field teamshelped in this process. The field team at eachpilot site was kept at a minimum size of five, witha combination of a formal training background insocial science and agricultural engineering. Thedisciplinary combination helped in handling thestrong socio-technical linkage that characterizeswater resources management for irrigatedagriculture, whereas, the small-size field teamswere successful in maintaining close interactionswith the community. The small size of the teamalso meant easy replicability of the catalytic efforton a wider scale.

The training given to the Social OrganizationField Teams3 on project objectives andmethodologies became an asset when they had to

through a common property regime in either ofthese cases can arrest the inequitabledistribution and overexploitation trends (Bromley1992; Ostrom 1990).

Many people, both within and outside the country,asserted that organizing water users for distribu-tary level management in Pakistan was a verydifficult task; some believed that it was impossible(Nasir 1992). Most of the contextual factorsdescribed in the earlier sections of this reportcontributed to this perception. Preliminary fieldinvestigations also indicated that organizing waterusers at the distributary level posed substantialproblems. Only some of the watercourses in thepilot area had experienced the formation of WUAssponsored by the On-Farm Water ManagementProgram, and these WUAs were already defunct.The water users in these watercourses wereparticularly hostile to the idea of yet anotherattempt to “organize” them. People in the areaappeared to be overwhelmed by problems ofsalinity and unproductive farming, and showedlittle patience to explore possible long-termsolutions. Specially designed social organizationmethodologies were helpful in this context.

The given social context demanded a fairlycautious and slow process of social organization,carefully designed with adequate trust-buildingstrategies, whereas, the physical context of thelarge contiguous canal irrigation system called fordesigning a method of appropriately sharing themanagement responsibility. Based on these twomain requirements, four special methodologicalfeatures emerged in this action research: 1)deployment of small social organization fieldteams consisting of persons with a strong local

3The term “SOFTware” was coined by IWMI’s staff in the Hakra 4-R Distributary pilot project in Punjab to distinguish these SocialOrganization Field Teams from IWMI’s other teams engaged in more technical work. More details of the organization process are givenin Bandaragoda et al. 1997.

Contextually Appropriate Methodologies

8

reflect the special features of the project design intheir field work. A fair understanding of theinstitutional implications of irrigated agriculture wasconsidered important in motivating people to seethe value of social organization. This training alsohelped the staff in undertaking self-assessment oftheir field operations and collective action.

Community-Based Social OrganizationVolunteers

To supplement the small field teams, a strategywas adopted to use community-based volunteersin social organization work. Initially, the projectdecided to call these volunteers “contact farmers”because they had to play a pivotal role as acontact between the field teams and thecommunity. Selecting some suitable persons fromthe local community to be deployed as “contactfarmers” was an important strategy in the socialorganization process. The term “contact farmers”was found to be associated with the “influentials,”big landowners and farmer leaders of the Trainingand Visit (T&V) system adopted by theAgricultural Extension Directorate. Since thefarmers in the study areas showed littleappreciation of the use of these elitist “contactfarmers” in the T&V system, and as theircontribution had not resulted in the properfunctioning of the T&V system, the term had anunfavorable connotation. In order to avoid farmers’mistrust from the start, the term “contact farmers”was replaced by the term “Social OrganizationVolunteers” (SOVs).

Training as a Motivating Influence

Experience of many infrastructure developmentprograms in Pakistan (Byrnes 1992) supported theidea of “putting people first” (Cernea 1985). Duringthe reconnaissance surveys in the pilot projectarea, many water users inquired about thepackage of physical incentives planned for the

project. They were accustomed to the governmentsubsidies on watercourse lining and tube-welldevelopment, etc. A considerable effort was spentto convince the water users of the need to getorganized first so that a form of collective actioncould benefit more from whatever the governmentcould deliver, or from their own resourcemobilization initiatives. They were eventuallyconvinced of this approach towards self-reliance.

The project did not have access to fundsallocated for any physical improvements to beeffected in the pilot sites. This was a majordeviation from the usual social organization projectdesigns adopted earlier in Pakistan and elsewherein the region. Both On-Farm Water Managementand Command Water Management programs hadphysical improvements as the main task, and theassociated institutional development componentwas to enable this primary task. Consequently,both programs could not achieve any meaningfulresults from the second objective (Byrnes 1992;World Bank 1996; Zaman and Bandaragoda 1996).Instead of physical incentives, this actionresearch program planned to use training as anincentive for organizational interactions. Theparticipatory approach adopted throughout theproject period typically suited this strategy, and aseries of training programs and similar interactionprograms were helpful in maintaining a steadylevel of enthusiasm among the water users.

The motivational effort, through training andinformation sharing, was also to engage the waterusers in building awareness, confidence, andmutual trust. There were no monetary incentivesand no promises of physical assets. The traininginputs were incorporated into other socialorganization activities, while monitoring the effectof each step and building on it. The strategy wasalso to share project-related information with thewater users in frequent meetings with small andlarge groups, in places considered as “neutral,”such as schools, mosques, playgrounds, andother community meeting places. Water usersshowed a greater interest in learning about thephysical aspects of the irrigation and drainage

9

systems than about proposed organizations. Thiswas quite natural as the physical subsystem of anirrigation system would be foremost in the mindsof the people. They liked to hear about quantityand quality of water they received, sedimentationin their canal system, and the conditions of thestructures. They were also keen to know aboutnew crop varieties, agricultural inputs, and moderncultivation methods for conserving water.

Phased Social Organization Process

Experiences in other countries suggested that“getting the process right” (Uphoff 1986) was avaluable initial investment in social organizationwork. An attempt was made, therefore, to developan appropriate process before embarking on theactual field work in the pilot project. A four-phaseorganizational development process4 was used asa guide during action research implementation.The four phases of this process are:

� support mobilization� initial organization� organization consolidation� organizational action

The support mobilization phase was a “get-set” stage during which the field teams weremobilized and trained, initial collaborativearrangements were discussed with the staff of On-Farm Water Management Directorate of theProvincial Agriculture Department, ProvincialIrrigation Department and other irrigation-relatedagencies, selection of the pilot sites was finalized,members for a field-level coordination committeewere identified, and initial baseline information wascollected.

In the second phase (initial organizationphase), some progressively advancing steps ininteracting with the community were taken. The

core social organization field activity wasimplemented during this phase. Starting from afamiliarization program, the field teams and thesocial organization volunteers proceeded throughthree other series of interactions, and finallyreached the culminating step of forming the wateruser federations.

The organizational consolidation phaseincluded a series of capacity-building programs toprovide WUO leaders and their members with thenecessary knowledge and skills to engage inactual water resources management tasks.Registration of WUOs with the On-Farm WaterManagement Directorate under the Water UsersOrdinance, and the preparation of jointmanagement agreements (JMAs) between thedistributary level water user federations and theProvincial Irrigation Department were two importanttasks undertaken during this phase. The lastphase is meant to give effect to the provisions ofthe joint management agreements.

In this organizational development process,many actors would need to contribute. A designteam coordinated the planning effort, and collabo-rated with the social organization field teamlocated in the three pilot sites. The selectedSOVs and a Field Implementation CoordinationCommittee (FICC) were the other partners in thefield. The FICC consisted of representatives fromvarious agencies providing irrigated agricultureservices to the farmers, including the civil admin-istration, and selected farmer representatives.Social organization activities were supported bysome collaborative activities by various agencies.The idea of conducting collaborative activities wasto maintain the water users’ interest on the actionresearch program. A facilitator, such as IWMI,would play the role of a catalyst in bringingvarious line agencies and other service deliverygroups to the water users on their request.

The overall process described above isdepicted in figure 2.

4This four-phase process for water user organization activities in Pakistan was adapted from the Maintenance and Operation (M&O)guidelines provided by Skogerboe, Poudyal, and Shrestha (1993).

10

FIGURE 2.Social organization process.

tional design. Since the interactions were betweenthe catalysts and the water users in the form ofdialogues,5 the stages of this iterative process ofsocial organization were named “Five DialogicSteps” as indicated in figure 3 below.

First Dialogue: A series of “familiarizationmeetings” to get to know the area and the peoplein general, and to introduce the purpose of thefield team’s visit, the idea of the pilot project, andits proposed activities to any person met in thecommand areas.

Five Dialogic Steps

An important feature of the iterative process wasthe progressively enhanced interactions in a seriesof meetings with the water users, which culminatedin forming water user federations in the pilot areas.Adopting a step-wise approach, and building on thesteps already taken, the process advances towardsthe group behaving on mutual trust, sharing infor-mation, consulting for consensus, developingoptions, and implementing an appropriate organiza-

5For characteristics of dialogical communication, see Schrijvers 1995:24.

11

Second Dialogue: A series of “rapport-buildingmeetings” to meet with the identified SOVs and afew other water users in small groups. The mainpurpose was to explain the objectives, status, andprograms of IWMI and build up fellowship with theSOVs and their colleagues.

Third Dialogue: A series of “consultationmeetings” to consult as many water users as

possible for developing tentative plans forestablishing water user organizations. Themeetings were to be in groups larger than thoseconsidered for “rapport-building” meetings. Theseconsultation or planning meetings formed a crucialstep in the social organization process to ensurethat the water users knew the project objectivesclearly, and to follow up on earlier rapport-buildingmeetings for clarifying any misunderstandingsamong the people regarding the program.

Fourth Dialogue: A series of “selectionmeetings” for the purpose of discussing theprocess for selecting or electing organizationalleaders at the primary (watercourse) level. Afterclarifying the elements of a democratic method forthis purpose, meetings were held for eachwatercourse to select the organizational leaders.With wide publicity and extensive personalinteractions, an attempt was made to get themaximum number of water users in eachwatercourse to participate.

Fifth Dialogue: “federation meetings” to initiatethe identification of office bearers for the pilotwater user federations. During these interactions,the water users were encouraged to select thewatercourse nominees, who would form thegeneral body of the federation in each pilot area,and then proceed towards selecting the WUFleaders.

FIGURE 3.Five dialogic steps leading to the formation of a wateruser federation.

Results and Discussion

Deployment of Community-BasedVolunteers

The most difficult part of this action researchprogram was to gain entry into the community ofwater users. Through the first dialogic step offamiliarization meetings, a group of community-based volunteers was selected to facilitate thisinitial entry and the subsequent social organizationprocess. At Hakra 4-R Distributary site in Punjab,

the field teams interacted with 486 individual waterusers, and, on the basis of their recommendations,selected 158 SOVs. With a similar effort, 160SOVs were selected for the three pilot sites(Bareji, Dhoro Naro, and Heran distributaries) inthe Sindh Province. Table 1 gives some importantcharacteristics of the selected SOVs, ascompared with the primary groups from which theywere selected.

12

TABLE 1.Comparison of some socioeconomic characteristics of SOVs and primary group of water users.

Item H-4R Bareji Distributary Dhoro Naro Minor Heran Distributary

All* SOVs All SOVs All SOVs All SOVs

Total number (n) 3,494 158 354 48 504 50 718 62

Nonoperator landowners—as % of n 9 0 50 62 10 58 47 74

Owner-operators—as % of n 85 99 44 17 33 20 40 24

Tenants—as % of n 6 1 6 21 57 22 13 2

Literacy rate as % of n 38 76 28 83 33 66 46 90

>10 years’ education 5 45 4 23 2 24 27 44

*All = All water users.

Sources: Cheema, Mirza, Hassan, and Bandaragoda 1997; Memon, Hassan, and Bandaragoda 1997; Hassan, Mirza, and Bandaragoda1996; Bandaragoda and Memon 1997.

� Interventions could be routed through localpeople, causing little room for mistrust; andthe SOVs in turn could encourage a localinitiative.

� The SOVs could reach the community in thelarge pilot distributary command area fairlyquickly, partially overcoming the project’s timeconstraint.

� As SOVs were deployed on a voluntary basis,the method was more cost-effective than inengaging paid agency staff and could easilybe applied on a wider scale.

The greatest impact of the SOV’s contributionwas felt when they took upon themselves toexplain to the community the objectives of theproject at a very critical stage of the project. Atthe early stages of the fieldwork program, socialorganization activities related to farmerparticipation in management were seen by somepeople, who had a vested interest in retaining thestatus quo, as part of a hidden agenda sponsoredglobally by aid agencies and countries moreclosely associated with them. The action researchproject was seen as an alien effort exerted toachieve conspiratory objectives detrimental to

The action research design expected toselect SOVs from community members whowere adequately informed about the communityand its needs, prepared to assist the actionresearch process, and showing an ability to playa very useful and active role in building trustamong the people. This initial assumption wasfound to be correct. As expected, the selectedSOVs knew (and were known to) the peoplefairly intimately, and the criteria of education andknowledge about the community supersededother features such as landownership, tenurialstatus, and age. The SOVs, being all based inthe community, were able to share thecommunity’s language, beliefs, traditions, andrituals, and were well aware of their commonneeds and problems, and became an acceptedgroup of opinion leaders.

Table 1 shows that the tenurial pattern in eachpilot site is reflected in the selection of SOVs. InHakra 4-R Distributary of Punjab, the majority ofthe water users are the owner-operators, whereas,in the three Sindh sites, there are a considerablenumber of nonoperator landowners, who employagricultural labor or tenants in operating theirlandholdings.

The methodology of using local volunteers hadthe following advantages:

13

Pakistan. At this stage, only the SOVs’ voluntaryactions helped to dispel such doubts andmisconceptions. The community preferred to relyon assurances and explanations of their ownopinion leaders.

Dialogic Process

The slow step-wise approach made the fieldteam’s task much more difficult than thetraditional approach of “handed down” instructions.The real challenge was that each step takencollectively with the people had to be based onpopular agreement on the previous step’s results.Gradually, the majority of the water users wereconvinced that the pilot projects were for their ownbenefit, but something they had to work hard tobuild by themselves. This effort was not withoutmisunderstandings and objections. The challengeitself provided a motivation to the socialorganizers and participant water users. It was avaluable experience for the field team members tosee how some of the water users played the roleof promoters of WUOs to argue with and convincetheir own fellow water users who were showingdissent.

Sometimes, the efforts of the field team intrying to forge some agreement and someconfidence among the people were followed byextremely frustrating negative results. One suchsignificant obstacle was the widespread rumor

originated by a few persons with vested intereststhat the action research program was a ployadopted by the donors and the government toincrease water charges. Often, the strategyadopted to overcome such constraints was togenerate internal discussions in small groups withthe help of SOVs and allow time for a consensusto be developed among the water users, on thebasis of information collected by themselves.Table 2 illustrates how the participation rategradually improved with the increased intensityand formality of social organization interactions,which progressed from very informal rapport-building meetings (second dialogic step) to veryformal federation meetings (last dialogic step).

Identification of Organizational Leaders

The action research design generated four mainprocedural emphases for selecting organizationalleaders for each level of the organizationalstructure:

1. Organization activities were conductedessentially in a participatory mode.

2. Equal opportunity was provided to all waterusers to participate.

3. Selection of organizational leaders waseffected in a democratic way.

4. Selection was done on the basis ofconsensus, and not on open competition.

TABLE 2.Participation rates* at different dialogic steps.

Pilot site Rapport-building Consultation Selection Federationmeetings meetings meetings meetings

Hakra 4-R 10 (n=3,494) 40 (n=3,494) 76 (n=3,494) 96 (n=25)

Bareji 29 (n=354) 58 (n=354) 72 (n=354) 90 (n=48)

Dhoro Naro 14 (n=504) 50 (n=504) 53 (n=504) 80 (n=50)

Heran 12 (n=1,076)** 51 (n=1,076) 64 (n=1,076) 96 (n=62)

*Number of persons who participated as a percentage of the maximum number (n) of persons expected to attend the respective DialogicStep. ** This increased number, as compared with the number given in table 1, includes some tenants who had since been admitted aslegitimate water users in Heran Distributary.

14

The water users, who were recognized by thecommunity of the command area in terms of legallyaccepted warabandi lists, formed the base groupfrom which all of the leaders were selected fordifferent organizational levels. These lists reflecteda more authentic situation regarding the eligibility ofpersons for membership in water user organiza-tions, and the numbers were slightly different fromthe total population identified during the baselinesurveys conducted in pilot areas. Since the meth-odologies of selecting organizational leaders inPunjab differed from those deployed in the SindhProvince, the two sets of data are analyzedseparately in the following sections.

Leaders at Hakra 4-R Distributary

At the Hakra 4-R Distributary site in the Punjab,the selection of four groups of organizationalleaders was progressively effected. For threedifferent levels of a hierarchy of organizations,namely, (1) the watercourse,6 (2) a subsystemconsisting of a cluster of watercourses, and (3)the distributary consisting of a number of suchsubsystems, the following four groups oforganizational leaders were selected:

� Watercourse nominees selected by the waterusers of the respective watercourse on thebasis of one nominee from each watercourse,as members of a Subsystem Water UserOrganization (WUO) for a defined cluster ofwatercourses.

� Subsystem WUO office bearers, selected bythe members of the Subsystem WUOs.

� WUF members, also selected by theSubsystem WUO members, on the basis offive selectees from each of the fivesubsystems.

� WUF office bearers selected by the 25 WUFmembers.

Social Characteristics

Table 3 presents a summary of the main socialcharacteristics of the organizational leadersselected in this process at Hakra 4-R Distributarypilot site, compared with the averages for thebase group of water users.

The average experience in irrigated agricultureamong the selected organizational leaders seems

6Watercourse is the tertiary level distribution canal providing water to nakkas (farm outlets), whereas, distributary is the secondary leveldistribution canal providing water to moghas (watercourse outlets). The distributary takes off from a main canal or a branch of a maincanal (see figure 4).

TABLE 3.Main social characteristics of organizational leaders at different levels of the Hakra 4-R distributary pilot site.

Group Average age Average experience Average number of(years) in agriculture (years) years of schooling

All water users (n=3,494) 49 28 3.3

Watercourse-level nominees (n=120) for 46 27 8.6subsystem WUO membership (15-90) (0-70) (0-16)

Subsystem WUO office bearers (n=25) 47 28 9.8(28-79) (5-60) (1-16)

WUF members (n=25) 41 23 11.1(25-62) (8-45) (8-16)

WUF office bearers (n=5) 46 28 10.2(38-62) (20-45) (8-12)

Note: Minimum and maximum values are given in parentheses.

15

to be consistently high, a factor that seems tohave mainly guided the water users in theirchoices. A similar high value appears to havebeen given to the educational level of theorganizational leaders at various levels, most ofwhom appear to be having a reasonable level ofeducation (the average being 9 to 11 years ofschooling).

Although some young persons have beenidentified as nominees for leadership at thewatercourse level, more mature, moreexperienced, and more educated persons havebeen selected for higher-level leadership positions.

Pattern of Landownership

Table 4 shows the pattern of landownership amongthe selected organizational leaders at differentlevels, in terms of both the land owned andoperated within the watercourse from where thenominee was selected and the total land ownedwithin and outside the Hakra 4-R Distributary. Inrural areas, landownership is considered animportant indicator of the influence a person canexert on others.

Table 4 also shows that the average extent ofland owned by the watercourse-level nominees asa total group is higher than that of the selectedSubsystem WUO office bearers, but is less thanthe averages for WUF members and office bearersselected by them. This shows that, once

watercourse level leaders were identified, therehad been a tendency towards selecting personsowning more land for positions of office bearers athigher organizational levels.

At a glance, it appears that the WUFmembers are the larger landowners. However, outof the 10 watercourse nominees owning 40hectares or more each (table 5), none wasselected as a WUF office bearer; only three wereselected as WUF members; and only two assubsystem office bearers. The larger landownersselected for the various offices also have otherqualifications, such as education and workingexperience. The 5 persons selected as WUFoffice bearers from the 12 nominees are medium-size landowners, their landownership ranging from17.4 to 38.6 hectares each. Even among theordinary WUF members, 14 persons own less than24 hectares each, and only 3 persons own morethan 80 hectares each.

Table 5 shows that a large proportion of the120 watercourse nominees is within the 4-40hectare landownership category. This distributionis fairly compatible with the distribution of theoffice bearers as well.

A large majority (92%) of the selectedorganizational leaders own less than 40 hectareseach. However, the other 8 percent of the leaderswho own more than 40 hectares each account for44 percent of the total land belonging to the wholegroup. These statistics depict a highly skewed

TABLE 4.Pattern of landownership among the selected organizational leaders.

Group Land owned at Operated area at Total land owned atparticular watercourse particular watercourse Hakra 4-R Distributary

(hectares) (hectares) (hectares)

All water users (n=3,494) 5.7 10.28 8.4

Watercourse-level nominees (n=120) forsubsystem WUO membership 11.4 8.8 16.3

Subsystem WUO office bearers (n=25) 8.8 7.9 14.1

WUF members (n=25) 15.3 16.1 21.1

WUF office bearers (n=5) 12.0 7.2 26.6

16

TABLE 5.Landownership among the organizational leaders of the Hakra 4-R Distributary pilot site.

Land size category Number of Percent of Total extent of Percent of Mean extent ofpersons total land (ha) total land land owned

<1 ha 3 2.5 0.6 0.02 0.20

>1 to 4 ha 16 13.3 38.1 1.5 2.38

>4 to 10 ha 33 27.5 219.6 8.7 6.65

>10 to 40 ha 58 48.4 11,159.8 46 20.0

>40 to 100 ha 6 5.0 431.6 17.1 71.93

>100 ha 4 3.3 674.0 26.7 168.50

Total 120 100.0 2,523.7 100.0 21.03

land distribution pattern. Given this skewed landdistribution pattern, an interesting aspect ofparticipation is that both the small landowners,and the big landowners, have shown an equalinterest in forming WUOs. Both categories ofpeople have a stake in joining the organizations;the small farmers wish to claim their share ofwater, whereas, the big farmers would like tosafeguard their current favorable situation. Thepotential of these new WUOs would largelydepend on their ability to coordinate these differentexpectations.

Organizational Leaders of Sindh PilotSites

In the three Sindh sites, each watercourse groupnominated two persons as members of thedistributary level WUF directly, bypassing the needfor an intermediary subsystem level WUO. Thedecisions on the organizational structures weretaken by the water users themselves in theirconsultation meetings, and they reflected rationalthinking, as the three Sindh secondary canal pilotsites were much smaller than Hakra 4-RDistributary in the Punjab. Their maincharacteristics can be seen in table 6.

In terms of landownership, the selectedorganizational leaders for the 80 WUAs in the

Sindh sites reflect the general characteristics ofthe water user community in the area. Thebaseline survey showed that the majority of thewater users were small landowners (averageextent owned by a water user was found to beonly 2 ha in Bareji, 3.5 ha in Dhoro Naro, and 4ha in Heran). However, in terms of theireducational level, the pattern contrasts with thecommunity distribution; table 6 shows that despitethe large proportion of illiterate people in thecommunity (64% according to the baselinesurvey), only a few uneducated persons havebeen chosen as WUA leaders.

Understandably, the water user communityhas included only a very small percentage oftenants and lessees among their selected leaders.A special reason for this collective decision isthe temporary nature of the tenurial relationships inthe Sindh Province. Tenants are often changedon a regular basis, and consequently, thetenants themselves show little interest inundertaking responsibilities in water resourcesmanagement.

Demonstrated Ability for CollectiveAction

The actions so far taken by the new water userorganizations are mainly in seven areas:

17

� distributary maintenance� preparations for more equitable water

distribution� establishing offices and bank accounts� negotiations with government authorities� agreements with input suppliers and

purchasers of produce� developing consensus among the

members for these actions� membership drive

Organized Action in Distributary Maintenance

The water user organizations mobilized sufficientresources to undertake desilting of their respectivedistributaries at all the pilot sites. During thisorganized desilting operation, the water usersdemonstrated much enthusiasm and cooperationamong themselves. The office bearers monitoredand coordinated the desilting process, allocated

the financial as well as human resources, andorganized the work equitably among the variouswatercourses. An evaluation of this work reportedthat a significant feature of this activity was theunprecedented organized action in mobilizingresources and attending to a well-preparedmaintenance plan. Water users assigned variousparts of the command area to each office bearerfor collecting an equal amount of dues per wateruser. There was wide publicity for the process ofcollecting dues so that there were no free riders.The WUOs, through discussions in their variousmeetings, identified some critical infrastructuredevelopment work items that would improve thephysical system in their respective distributaries.The Irrigation Department gave official approvalbefore the WUOs undertook the physical work,mostly with their own resources. The projectprovided technical assistance and some funds forprocurement of cement and bricks. Table 7 shows

TABLE 6.Profile of WUA office bearers in the Sindh pilot sites, numbers, and percentages.

Landownership Bareji Dhoro Naro Herann = 116 n = 195 n = 239

0 – 10 hectares 72 (62.1) 136 (69.8) 145 (60.7)

>10 – 20 hectares 15 (12.9) 34 (17.4) 82 (34.3)

>20 – 30 hectares 13 (11.2) 9 (04.6) 5 (02.1)

>30 – 40 hectares 3 (02.6) 2 (01.0) 5 (02.1)

>40 hectares 13 (11.2) 14 (07.2) 2 (00.8)

Tenancy status Bareji Dhoro Naro Herann = 116 n = 195 n = 239

Landowners 73 (62.9) 94 (22.1) 102 (42.7)

Owner-operators 24 (20.7) 68 (32.9) 106 (44.3)

Lessees and tenants 8 (07.0) 17 (08.8) 30 (12.6)

Kamdars (managers) 11 (09.5) 16 (08.2) 1 (00.4)

Educational status Bareji Dhoro Naro Herann = 116 n = 195 n = 239

Illiterate 17 (14.6) 43 (22.1) 35 (14.6)

Primary 53 (45.7) 79 (40.5) 42 (17.6)

Matric 22 (19.0) 43 (22.0) 85 (35.8)

Intermediate and above 24 (20.7) 30 (15.4) 77 (32.2)

Note: Percentages are given in parentheses.

18

the assessed value of the items of workaccomplished through collective action during1997 and 1998 canal closure periods.

The socio-technical linkage in collective actionhas been a key element in the design of thisaction research. The willingness of the waterusers to gain an understanding of their technicalsystem encouraged them to get organized formanaging the system, and for engaging incollective maintenance programs aimed atimproving the productivity of water use. The intentof a maintenance program to fix all flow controlstructures required for the improved hydraulicoperation of the system is to ensure that they canfunction both as flow control and flowmeasurement structures. In this sense,maintenance is a support activity to facilitatecanal operations (Skogerboe, Paudyal, andShrestha 1993). The information gathered throughtraining programs covered the water supplysituation, water distribution pattern, irrigationstructures, channel reaches, channel physicalconditions, existing maintenance strategies,involvement of beneficiaries, government agencyrole, and possible constraints. After acquiring agood knowledge of maintenance needs,

maintenance requirements could be categorizedeither as essential structural maintenance, ordeferred maintenance.

In addition to the maintenance training, thewater user groups also participated in activities toacquire basic ideas about operating the physicalsystem to distribute water equitably. Improvedirrigation and agricultural practices formed thefocus of a number of field visits to experimentalstations and the demonstration plots establishedwithin the pilot areas. Diagnostic “walk-thru”maintenance surveys, which were conducted tolist in detail all deferred maintenance needs alongthe distributary and minors, served to achieve thefollowing main objectives:

1. Water users gained an understanding aboutthe technical aspects of maintenanceproblems.

2. Field teams were able to benefit from farmers’views about maintenance problems.

3. Both groups collectively appreciated historicaland social aspects of specific watermanagement difficulties.

TABLE 7.An assessment of expenditure for different activities by the water users of the pilot distributaries in rupees(US$1.00=Rs 40).

Activities Water User Federations of Pilot Distributaries

Hakra 4-R Dhoro Naro Heran BarejiDistributary, Minor, Distributary, Distributary,Haroonabad Nawabshah Sanghar Mirpurkhas

Desilting 1997 n.a. 67,500 109,000 6,800

Desilting 1998 124,000 25,700 92,000 56,000

Development work n.a. 164,600 148,900 95,500

Repair of head regulator n.a. n.a. 5,000 n.a.

Construction of WUF office n.a. 25,000 n.a. n.a.

Repair of WUF office 10,000 n.a. 6,000 1,500

Total assessed value 134,000 282,800 360,900 159,800

Source: Zaman 1998 and Project Progress Reports.

19

An Attempt to Improve Equity

The WUF of Dhoro Naro Minor attempted tomodify outlet structures to ensure the delivery ofwater according to the agreed water rights and therelated water allocation pattern. This work had tobe suspended when the Joint ManagementAgreement between the WUF and the ProvincialIrrigation Department (PID) was not ratified byhigher authorities. However, the fact that the WUFwas ready to undertake a change reflects thepotential for effective M&O through collectiveaction.

Short-Term Impacts

After the formation of WUFs, the water usersreported that the frequency of breaches in theDhoro Naro Distributary declined by about 50percent. One reason could be that some of thebreaches that occurred in the past were probablyman-made. The organized efforts in stabilizing andstrengthening the weak points of the canal banksof the pilot distributaries also could havecontributed to this reduction in incidents ofbreaches. The water users generally refer to someimprovement in the water flow in the tail reachesin the distributaries, and attribute it to theorganized desilting campaigns. The sharing ofinformation among the water users has beenenhanced by regular interactions in the WUOmeetings.

Potential for OrganizationalConsolidation

Institutionalizing an organization for collectiveaction is based on a nested set of rules. Externalrules associated with institutional support for itslegal recognition would lead to the organization’sown rules of conduct (or rules for makingcollective choices), which in turn would determinethe operational rules on how resourcemanagement is to be implemented. These

“decision-making arrangements” (Oakerson1992:46) largely determine the organization’scapacity to proceed with the management tasks.Viability of an organization depends on its strengthto make internal rules (i.e., rules of conduct andoperational rules mentioned above), and to applythose rules effectively. Both these functions,however, depend largely on the strength of theexternal rule system, or the organization’s legalenvironment.

With the enactment of Provincial Irrigation andDrainage Authority Acts (in Punjab on 2 July1997, and in Sindh on 15 September 1997), themajor part of the external rule base is in place.Regulations to these acts, once finalized, wouldclarify further the procedure for internal rules forWUOs. The pilot projects have preempted thisstep by developing bylaws for each pilot siteorganization, and entering into a negotiationprocess with the government authorities to finalizethe bases for shared management of waterresources. However, for a WUO to be aninstitution, it has to persist and develop to thepoint where it is commonly perceived as valuableand useful (Merrey 1993). As has beenexperienced in the earlier efforts in forming WUAsprimarily for watercourse improvement in Pakistan,an organization formed for short-term objectives ismerely a structure of identified roles accomplishedby a few selected individuals, such as thepresident, secretary, and treasurer. At best, it canbe referred to as an “organization,” but often it islimited to an ad-hoc group of a few people workingtogether to achieve a temporary task. Once theshort-term objectives are achieved, suchorganizations can, and usually do, disappear.

The task of managing water resources in acommon pool water resource system (e.g.,distributary) through collective action gives thenew WUFs and their member WUAs a morepermanent objective for organization.Implementation of agreed water rights and otherrelated bylaws for reliable and equitable waterdistribution among watercourses and memberwater users is the most significant management

20

task that will engage the WUFs on a continuingbasis. Other similar persistent responsibilitieswould include resource mobilization forimplementing recurrent management tasks,dispute resolution, interacting with outside groupssuch as the government and private sectororganizations, seasonal planning for optimum cropproduction, managing the proper use of inputs,and profitable procurement and marketing efforts.

The Joint Management Agreements (JMAs)developed by the WUFs and presented to thegovernment, though not formally signed bygovernment authorities yet, reflect a commonunderstanding between the water user groups andgovernment agencies that there will be clearlydefined water rights for WUFs beyond thedistributary head regulator up to moghas, andthrough the involvement of their member WUAs,up to nakkas. These JMAs specify responsibilitiesand rights of both WUFs and governmentauthorities. As the water users in pilot sites hadfelt that the water supply to these particular siteswas purposely reduced by local irrigation staff, theJMA requires from the government an assuredsupply, “calculated on the basis of the averagedeliveries for the past two to three years.”

While continuing to negotiate with thegovernment for having the JMAs formalized, theWUFs have engaged in a number of otheractivities aimed at organizational consolidation,such as developing draft bylaws, holding regularmeetings, and taking collective action for resourcemobilization, capacity building, and disputeresolution.

Shared Management through aCombination of Property-RightsRegimes

Once the organized water users take over thesecondary level canal subsystems, the waterresource system in the distributary can be treatedas a common pool resource, which is owned andmanaged as common property. So far in the Sindh

Province, through the JMAs developed by theSindh pilot sites and accepted on principle by thegovernment on 16 October 1997, theunderstanding is that “the right of use on thedistributary, inclusive of all structures, will behanded over to the Water Users Federation formanagement.” Due to a procedural difficultycaused by a delay in publishing rules andregulations under the Sindh Irrigation and DrainageAuthority Act, the JMAs could not be legallyformalized. In the Punjab Province, negotiationsbetween the Hakra 4-R Distributary WUF and theProvincial Irrigation Department have succeeded inan agreement for a limited transfer of managementresponsibilities, but its formalization has also beendelayed until PIDA regulations are finalized.

In both provinces, a clear understanding hasbeen reached that the Area Water Boards (AWBs)will manage the main canals, leaving thedistributaries to the WUFs. Initially, the AWB is tobe constituted with government-appointed boardmembers, although some of them will necessarilybe farmer representatives as specified in the PIDAlaw. To this extent, the main canal and its waterwould basically be state property until such timethe AWB becomes a genuinely farmerrepresentative body selected by the water usersthemselves. In the latter event, the part of thewater resource system up to the head of the maincanal can become common property.

In the present circumstances, as the actionresearch design envisaged, the canal systemabove the head of the distributary would be stateproperty, and the distributary and its water wouldbecome common property. Decisions on waterrights and water distribution patterns within thedistributary would largely be determinedcollectively by the WUOs. Interactions with thewater users showed that the most significantmotivating factor that tends to encourage theindividual water users to join organizations is thispotential for collective action to own and managewater resources in the distributary. The enhancedcapacity to negotiate with the government agencyregarding inter-distributary rotation schedules,

21

reliability of water supply, and upstreammaintenance activities was cited as anadvancement in local management. Similarly, theincreased capability of the larger collectivizedgroup—the WUF owning the common property ofthe distributary—for identifying and minimizinginstances of free riding and rent-seeking, whichare normally associated at the level ofwatercourse groups, was also highlighted as anadvantage.

The water resources in a watercourse for agiven warabandi turn remains as private propertyof the individual water user having that turn.Figure 4 gives the present and proposed property-rights regimes for a canal water resource system.

In a combination of property-rights regimes(state, common, and private), the interfacesbetween different regimes become important lociof complex human interactions. These interactionsbetween different regimes are determined by setsof external and internal rules. Converting thedistributary level water resource system from itspresent “state property” status to a “commonproperty” regime, the farmer-bureaucracy interfaceis shifted to the head regulator of the distributarycanal. While the external rules would largelydetermine the interactions at this interface,appropriate internal rules would determine themanagement of common pool water resources inthe distributary. Either set of these rules will bethe result of a series of negotiations between thewater users and the government.

The private property regime associated withwater resources in the watercourse and itsapplication in crop fields also requires that thehandling of surplus water is a private responsibilityuntil it reaches a drainage facility commonlyowned by a WUO. The water users have alreadyacknowledged their readiness to own thisresponsibility, and also indicated the possibility ofthe WUF undertaking drainage responsibility withinthe distributary command area, until the collected

surplus water reaches a larger drainage facilityoperated by a state agency.

In this combination of property rights regimes,a significant effort is necessary to effectivelyreorient roles of the state, community, and theindividual. At the top level of the canal system,the state will be responsible not only for themanagement of reservoirs and main canals, butalso for determining higher-level allocation ruleswith greater clarity. At the distributary level, thecommunities will adhere to the allocated water andtake the responsibility for the equitable distributionof water according to their own agreed internalrules, and also for mobilizing resources formanaging the distributary as well as sharing theupstream canal management cost with the state.Much greater freedom can be exercised by theindividual water user at the watercourse level,once their water allocations are known.

Figure 5, which is an adaptation (Chambers1988:36) depicts the interfaces between waterusers, their organizations, and the state.Appropriately formulated external and internal rulesfor this new institutional arrangement would guidethe functioning of these various interfaces. Thisaction research program envisaged a series ofnegotiations between the government and thewater users (at least the pilot WUOs) indeveloping external and internal rules. However,some delays in identifying the desired institutionalframework for managing water in the main canaland above became a constraint to this desiredprocess. Until the closure of the present phase ofthe action research program, no Area Water Board(AWB) was established in the canal system. Thenew Provincial Irrigation and Drainage Authoritylaws do not provide for fully farmer-managedsystems at the main canal level, as the AWBsare to be formed out of a combination ofprofessionals and some farmers, both categoriesappointed by the government.

22

FIGURE 4.Property rights of a water resources system.

23

FIGURE 5.Shared water resources management (WRM)* in large canals.

*WRM covers O&M of physical systems, water use and disposal, and related environmental aspects.

24

The action research has shown that, withadequate opportunity to freely interact amongthemselves, the water users in the pilot sites arecapable of selecting their organizational leaders ina democratic manner to include representationfrom all sections of the community.

Methodologically, the deployment ofcommunity-based volunteers and the strategy ofdeviating from the usual system managementemphases and government budgetary imperativesfor social organization paid dividends. Theprolonged confidence-building efforts adopted in adialogic approach helped in improving the trust,not only between the community and the changeagents, but also among the water usersthemselves. In a participatory action researchmode, the water users in the pilot areas engagedthemselves in a series of consultation meetings toidentify their needs for collective action, plannedthe needed organizational mechanisms,established primary and secondary level usergroups, and federated them in a composite,representative organization. They proceeded totest their organizational capacity by successfullymobilizing resources, collectively planning urgentmaintenance activities, and implementing aprogram of action during the canal closure period.In this process, they also identified means ofimproving equitable water distribution within theirdistributary canal command area and resolvedpreliminary conflicts confronted in this collectiveaction process. Social organization field teamsplayed a facilitating role, with occasional attemptsat some elements of social engineering.

One constraint has been the reluctantparticipation by the bureaucracy in this institutionaldevelopment effort. This is not surprising. In manycountries where social experiments have beenattempted, the main resistance to change hascome from the bureaucracy. Despite this

constraint, the satisfactory situation involvingthree other main actors (community, opinionleaders, and political leaders) could be identified inthe following preliminary findings:

1. Initially, the members of the community in allof the pilot project areas were reluctant toparticipate in any form of interaction with thefield teams. This initial diffidence wastransformed to a gradual appreciation of self-management concepts and the need forgetting organized. Once organized, theysuccessfully tried some maintenance andequity improvement strategies. With theconfidence gained, they are now keen to takeover the distributaries from the government asfully farmer-managed systems.

2. Socially differentiated groups, such as head-end and tail-end water users, large and smalllandowners, landowners and tenants, andinfluentials and vulnerable groups, have allparticipated in forming the WUOs. Ademocratic process for selecting the WUOleaders was successfully tested.

3. Political leaders have provided a nonpartisansponsorship without interfering in this socialorganization process. The acceptance of thenew WUOs by the community is almostunanimous.

Usually, the fairly well-established ruralleadership ensures that radical changes in attitudedo not occur quickly among the ordinary membersin the community; the same leadership tends totake over any rural organization that is sponsoredby top-down approaches. However, this actionresearch shows that, given the equality ofopportunity and the democratic process for

Conclusion

25

selecting organizational leaders, possibilities existin increasing the extent of participation bytraditionally marginalized groups. The selection ofa majority of medium-size and small landowners,and even a few tenants as WUO leaders,avoidance of influence by competing politicalparties, and reasonably equitable considerations indecision making seen during a short period of timein this social organization action research indicatea good potential for avoiding the usual dominanceby large landowners.

One important area for continued researchremains in validating the economic gains fromcollective action through a combination ofproperty-rights regimes in the context of largeirrigation canals. The new pilot WUOs could noteffectively take over management responsibilitiesfrom the government as the enabling legalframework was not complete on time for theaction research. Consequently, they were unableto spend sufficient time in collective action, andfully test their ability for managing water resourcesfor increased productivity in irrigated agriculture asenvisaged. Yet, they have proceeded to build theircapacity for achieving this goal, should theopportunity be given to them in a not too distantfuture.

Another area for further research is to identifythe strategies with which a demand-driven bottom-up approach to establish decentralizedmechanisms for the management of water in largecanal systems can be institutionalized in thepresent sociopolitical culture in South Asia. More

specifically, what are the conditions under whichthe dominance of both the large landowners andthe state can be minimized to allow for genuineand sustained popular participation in resourcemanagement in the rural sector?

So far, there has been a slow but clearlydiscernible shift in the attitude of the electedpolitical leadership towards appreciating the valueof self-reliance that has been demonstrated by thenew WUOs. Whether this trend would continue isyet to be seen. Right now, it appears to beprompted by donor influence and the urgent needto mobilize resources for rural development. Inthis context, however, the willingness of thepolitical leadership to replicate pilot efforts andsome elements of the new approach on a widerscale in all the four provinces in Pakistan augurswell for the immediate future.

However, the action research efforts were notwithout the constraints of usual bureaucraticdelays and apathy towards change. The actionresearch program remains as an unfinishedagenda. The unprecedented enthusiasm generatedamong the water users through the bottom-upsocial organization approach in forming water userorganizations is confronted by the rigor oftraditional top-down planning processes forestablishing the Provincial Irrigation and DrainageAuthorities and Area Water Boards. Whether anenlightened political leadership can effectivelyneutralize the forces of traditional vested interestsagainst change in the rural sector is an openquestion.

26

The “crafting” of new irrigation institutions needs totake into account a number of contextual variablesthat contribute towards performance (Ostrom1992). While designing this action researchprogram, an effort was made to make itsobjectives and methodologies compatible with thecontext of Pakistan. A brief description of thiscontext is given below.

Physical System

Pakistan’s heavy investment in irrigationinfrastructure has given the country the world’slargest contiguous canal irrigation system. Themassive resource base of the Indus BasinIrrigation System is the cumulative effect of morethan a hundred years of consistent investment inirrigation development. The Indus Basin Project(IBP) of the 1960s alone saw an increase in thetotal water supply for irrigation from about 79billion cubic meters at the time of independence toalmost 135 billion cubic meters by the end of theIBP effort (Bandaragoda 1993:10).

Pakistan’s irrigation system, which is basedon the original objective of irrigating the maximumpossible area from the available water supplies, ischaracterized by the following main features:

� run-of-river water supplies� protective irrigation� low water allocations of 0.21-0.28 l/s/ha

(3-4 cusecs per 1,000 acres)� low cropping intensity (annual average

75%)� infrastructure designed for equity and

reliability of supply

� few gated structures and minimaloperational adjustment required

� proportional outlet structures drawingdesign discharge

Social System

The operation and maintenance of this extensiveirrigation system and its related drainage and floodprotection measures have largely been theresponsibility of the government. The ProvincialIrrigation Departments (PIDs) are responsible forthe major part of the task, which is the O&M ofthe main and secondary canals, while the farmersattend to the maintenance of the tertiary levelwatercourses.

A hierarchy of organizational units is involvedin the O&M of a typical canal system. AnExecutive Engineer playing a pivotal role in canaladministration is in charge of a canal Division,which is the executive unit for operationalactivities, and he functions under theadministrative control of a SuperintendingEngineer, who is the head of a Circle consisting oftwo or three Divisions. The SuperintendingEngineer of a Circle reports to the Chief Engineer,who oversees a number of such Circles. ADivision is further divided into three or fourSubdivisions, each headed by a SubdivisionalOfficer (SDO), who is also a qualified engineer. ASubdivision, ordinarily, consists of three or fourEngineering Sections and two to three Zilladari orRevenue Sections. The head of an EngineeringSection is a Sub-Engineer, who is responsible forthe distribution of supplies and the maintenance ofsecondary level discharge up to about 2 to 4

APPENDIX

The Context

27

cumecs (100 to 150 cusecs). The Sub-Engineer isassisted by Masons, Mistries, Mates and CanalPatrols for maintenance and watching of channels,and also has gauge readers for regulation andobservation of water flow. A Zilladari Section isheaded by a Zilladar who supervises the work ofabout 10 Patwaris (Irrigation Record Keepers),each Patwari being required to record the extent ofirrigation of 1,200 to 2,000 hectares.

The pattern of staff distribution deployed tooperate a canal system under the traditionaldesign has remained largely unchanged since theinception of the existing canal irrigationadministration in the late 1800s. As the systemwas designed for a low management intensity, thedensity of irrigation staff is lower in thesubcontinent than in many other irrigated areas.The average irrigation staff per 1,000 irrigatedhectares is in the order of 3 to 5 in many of thesecanals, compared to around 25 in South Koreancanals (Wade 1988). However, the performance ofthis static irrigation bureaucracy in thesubcontinent has been known to be onthe decline, and its character has beenbasically control-oriented rather than service-oriented.

A largely uncoordinated institutional frameworkcompounds this situation further. Federalresponsibility for resource allocation, provincialresponsibility for irrigation management, largeorganizations with centralized administration, largenumbers of water users with little involvement inirrigation management decisions, difficultcoordination among agencies and their subunitsand functions, numerous laws and proceduresmixed with traditional concepts and sporadicamendments by occasional enactments andpromulgations, and more importantly thecountervailing forces that act against formal rules,all contribute to the complexity of Pakistan’sirrigation institutions.

In sum, the following main institutional factorsaffecting irrigation performance in Pakistan can beidentified (Bandaragoda and Firdousi 1992):

� the overriding effect of socially evolvedinformal institutions over the formal rulesand management decisions

� the obsolescence of irrigation rules,codes, and procedures

� the declining relevance of organizationalstructures in the light of changedcircumstances

Socioeconomic Changes

With the political development followingindependence, the irrigated agriculture sceneunderwent some changes (Bandaragoda 1996).Populist approaches of a newly emergingdemocracy tended to bring about substantialinfluence on many aspects of canal administrationthat, through a cycle of mutually reinforcing socialfactors, led to the present situation of free ridingand rent-seeking. Most of the traditional designfeatures have outlived their usefulness in thecontext of these changed social conditions. Theideas of “protective” irrigation and equitable waterdistribution embodied in the early design criteriaare no longer readily applicable. Interactingchanges in the operational environment resulted inincreased indiscipline in the operation of thesystem, poor maintenance, demand for moreirrigation water and for its greater reliability,uncontrolled groundwater development, increase incropping intensities (over 100% in many systems),and diversified cropping patterns.

For instance, general deterioration of thephysical infrastructure, coupled with operationalirregularities, adversely affected the reliability ofirrigation water supplies, as well as the equity ofwater distribution within the system. As thenumber of small farms increased due tosubdivision and transfer of land, the emphasis ofobjectives in irrigated agriculture started to shiftfrom productivity per unit of water to productivityper unit of land. The cumulative effect of thesepolitical and socioeconomic changes can be seen

28

in the present social setting of Pakistan’sirrigation, which can be outlined by the followingmain features:

� skewed landownership pattern� increasing number of small landholders

due to fragmentation of land� highly centralized irrigation administration� lack of accountability� rampant rent-seeking behavior� political interference in administration of

the irrigation systems� inequity in water distribution

Constraints on O&M Budgets

For the projects before independence, during theFirst Five-Year Plan period (1955-60), the share ofpublic investment on agriculture and irrigation wasaround 30 percent. This share increased to about46 percent during the Second and Third Planperiods (1960-70) when the Indus Treaty projectswere implemented, but since then has declinedrapidly to a level of about 17 percent in the SixthPlan period in the 1980s (Hamid and Tims 1990).Although the share of the government budget fordevelopment work has gradually declined sincethe completion of the IBP, the governmentcontinued to allocate resources for upgrading andenhancing the system to meet an increasingdemand for irrigation.

The annual O&M allocations for the PIDsgradually became insufficient due to inflation andheavy inter-sector competition for resources, andthe management of O&M became increasinglyineffective due to changing socioeconomicconditions. Despite the increases in water charges(e.g., an average of 5% per annum from 1983 to1987), the revenues as a percentage of O&Mcosts declined (e.g., from 53% to 38% during theperiod). Increases in O&M costs, low assessmentof water charges, and low recovery rates, allcombined to form this imbalance (Water SectorInvestment Plan 1990).

With inadequate maintenance, the canalsystem started to deteriorate. The resultantdecline in performance deprived the country of itsexpected return on investment in irrigationdevelopment. Pakistan’s crop yields remaingenerally low, or have progressed only very slowlywhen compared to the crop yields in many othercountries. Similarly, poverty has stubbornlypersisted in rural areas despite their proximity toirrigation (Ministry of Food and Agriculture 1988).A daunting aspect of this unsatisfactory irrigationperformance, despite the favorable resource baseand related technological advances, could well bea substantial food deficit in the future, particularlyin view of the country’s fast-growing population.

Attempted Remedies

Donors and external evaluators started to drawattention to the need to identify correct solutionsto improve this state of unsatisfactoryperformance. Their proposition was that Pakistan’smajor irrigation problems were not due to watershortage. Therefore, instead of investing on furtherexpansion of the physical system, performanceimprovement could be achieved by introducing andsustaining appropriate institutional andmanagement innovations.

Illustrating this concern, the World Bank,United States Agency for InternationalDevelopment, and other donors focused theirattention on management rather than onconstruction. For example, the World Bank fundedfour projects: On-Farm Water ManagementProjects I and II (1981-1992), Irrigation SystemRehabilitation Project (1982-1987), and CommandWater Management Project (1984-1992) to addressthe major system management and institutionalissues. All these four projects, at an investmentlevel of US$175 million, were to concentrate onreducing drainage and saving water using existinginfrastructure, rather than building new dams.Further, each of these projects had a specificallydesigned institutional component. The importance

29

of this shift of emphasis was further accentuatedby continued donor pressure for institutionalreforms, which resulted in a slow movementtowards change. This situation is summarizedbelow.

1. Provincial legislation was passed in the early1980s, allowing the formation of Water UserAssociations (WUAs) on individualwatercourses. Since that time, thousands ofWUAs have been organized with governmentsubsidy and support given under the On-FarmWater Management (OFWM) developmentprogram.

2. In some selected command areas, a certaindegree of institutional coordination wasattempted under the Command WaterManagement Program (CWMP), in whichfarmer participation was made an essentialrequirement for project implementation.

3. Various evaluations of this work (e.g., Byrnes1992; Asrar-ul-Haq, Shahid, and Akram 1996)created the common impression that theseattempts at farmer involvement in themanagement of the irrigation system did notlead to sustained farmer participation or tolasting benefits. The World Bank’s post-projectevaluations later confirmed that the projectsachieved their physical components(watercourse water losses reduced from about40% to 25–30%; and annual water savingsfrom the four projects amounting to about 2.3billion cubic meters), but failed in most of theirinstitutional objectives. The evaluations furthercommented that the newly formed WUAs, tomeet project conditions, were merely tokenassociations or the old watercoursecommittees renamed, making the wholeexercise an “empty ritual” (World Bank 1996).

4. Government policy makers started toparticipate in discussions with the donors onpossible institutional reforms. Several

seminars were held among local opinionleaders to discuss the implications ofsuggested reforms, and these ideas were laterexpressed in published form (e.g., Asrar-ul-Haq, Shahid, and Akram 1996).

5. Meanwhile, the World Bank in their report on“Pakistan, Irrigation and Drainage: Issues andOptions” (World Bank 1994) proposed areorganization of the whole irrigation sector,including the establishment of autonomouspublic utilities for the management (includingoperation and maintenance) of the irrigationwater. Many government officials found thisapproach too radical, but recognized the needfor some institutional change.

6. An initial government agreement on the needto change was achieved at a seminar on“Participatory Irrigation Management,” co-sponsored by Pakistan’s Ministry of Waterand Power and the World Bank’s EconomicDevelopment Institute (EDI), which was heldin Islamabad during 2-6 October 1994. Thisinitiative was followed by another EDI-sponsored workshop held in Burban (a hillresort close to Islamabad) in October 1995,during which the representatives from fourprovinces worked out tentative action plans forinstitutional change.

7. In the midst of considerable pessimism aboutparticipatory irrigation management and itsvalidity in Pakistan’s large canal systems, aconsensus started to develop on the need toundertake some pilot projects in selectedlocations.

8. IIMI’s study results over the past decade inirrigation system management, policy, andinstitutional analysis in Pakistan coincidedwith, and have probably helped to catalyze,these newly emerging concerns and interests(Bhutta and Vander Velde 1992; Vander Veldeand Murray-Rust 1992; Bandaragoda and

30

Firdousi 1992; Bandaragoda 1993; Restrepo,Bandaragoda, and Strosser 1994;Bandaragoda and Saeed ur Rehman 1995).

9. Currently, there is a growing awarenessregarding the necessity for farmers’involvement in O&M, often prompted by donorconcerns, and also based on the realizationthat declining budgetary capacities werecontinuing to have adverse effects. Yet, thereis considerable pessimism among manygovernment officials about being able to formeffective farmer organizations and their impacton the productivity and sustainability ofirrigated agriculture.

Most of the contextual factors outlined aboveseemed to favor an increased involvement of thewater users in managing the present situation. Forinstance, by bringing management closer to theusers, the accountability for scarce waterresources was more likely to be improved.Additionally, by fostering maximum participation,the alliances made between some of the waterusers and the officials based on their vestedinterests were more likely to be minimized. Also,

collective action seemed to be the best way ofreducing irrigation misconduct among the waterusers themselves.

Although it has been a slow process ofchange, Pakistan’s current strategy for institutionalchange in this vitally important water sectorappears to have some practical and contextuallyappropriate elements. Sensing the initial objectionsto concepts, such as “privatization of irrigation”and “irrigation management turnover,” the plannersshifted to a strategy of organizational reform as aninitial step. While many other countries flounderedon this essential requirement, and generated verydisappointing results after their enthusiasticbeginnings on management transfer plans,Pakistan put forward the ideas of “decentralization”and “participatory irrigation management” toneutralize initial political objections. The reformsstarted with the enactment of new laws in theform of Provincial Irrigation and Drainage Authority(PIDA) Acts of 1997, and the appointment of PIDABoards. Even if these initial intentions were to beproved less than totally pure, the firm legalfoundation laid out through these preparatoryactivities would serve as a basis for local-levelawakening.

31

Literature Cited

Asrar-ul-Haq; B. A. Shahid; and U. Akram. 1996. Privatization and commercialization of irrigation management: Arethe new strategies workable in Pakistan’s setting? Paper No. 553. Proceedings of the National Engineering Con-gress. Lahore, Pakistan: Engineering Congress.

Bandaragoda, D. J. 1993. The role of research-supported irrigation policy in sustainable irrigated agriculture: An in-terpretive precis of the case of Pakistan. Colombo, Sri Lanka: International Irrigation Management Institute.

Bandaragoda, D. J. 1996. Institutional framework for irrigation: Some salient features of the South Asian situation (pa-per presented at the DSE/IIMI Workshop on Institutional Framework for Irrigation, Chiang Mai, Thailand, 1-6 No-vember 1993). In Institutional framework for irrigation: Proceedings of a workshop, ed. C. Abernethy, 39-53. Co-lombo, Sri Lanka: International Irrigation Management Institute.

Bandaragoda, D. J.; and G. R. Firdousi. 1992. Institutional factors affecting irrigation performance in Pakistan: Re-search and policy priorities. IIMI Country Paper - Pakistan No. 4. Colombo, Sri Lanka: International Irrigation Man-agement Institute.

Bandaragoda, D. J.; and Yameen Memon. 1997. Moving towards participatory irrigation management. IIMI’s Paki-stan National Program. Research Report No. R-26. Lahore, Pakistan: International Irrigation Management Insti-tute.

Bandaragoda, D. J.; and S. U. Rehman. 1995. Warabandi in Pakistan’s canal irrigation systems: Widening gap be-tween theory and practice. IIMI Country Paper - Pakistan No. 7. Colombo, Sri Lanka: International Irrigation Man-agement Institute.

Bandaragoda, D. J.; and Waheed uz Zaman. 1996. Government interventions in social organization for water resourcemanagement: Experience in a command water management project in the Punjab, Pakistan. IIMI’s Pakistan Na-tional Program, Research Report No. R-14. Lahore, Pakistan: International Irrigation Management Institute.

Bandaragoda, Don Jayatissa; Mehmood ul Hassan; Zafar Iqbal Mirza; Muhammad Asghar Cheema; and Waheeduz Zaman. 1997. Organizing water users for distributary management: Preliminary results from a pilot study in theHakra 4-R Distributary of the Eastern Sadiqia Canal System of Pakistan’s Punjab Province. IIMI’s Pakistan Na-tional Program. Research Report No. R-25. Lahore, Pakistan: International Irrigation Management Institute.

Bhutta, M. N.; and E. J. Vander Velde. 1992. Equity of water distribution along secondary canals in Punjab, Pakistan.Irrigation and Drainage Systems 6:161-177.

Bromley, D. W., ed. 1992. Making the commons work: Theory, practice and policy. San Francisco: Institute for Con-temporary Studies.

Byrnes, Kerry J. 1992. Water users associations in World Bank-assisted irrigation projects in Pakistan. World BankTechnical Paper No. 173. Washington D.C.: The World Bank.

Cernea, Michael. 1985. Putting people first: Sociological variables in rural development. New York: Oxford Univer-sity Press.

Chambers, R. 1988. Managing canal irrigation: Practical analysis from South Asia. Cambridge: Cambridge Univer-sity Press.

Cheema, Muhammad Asghar; Zafar Iqbal Mirza; Mehmood Ul Hassan; and Don Jayatissa Bandaragoda. 1997. So-cioeconomic baseline survey for a pilot project on water users organizations in the Hakra 4-R Distributary com-mand area, Punjab. IIMI’s Pakistan National Program. Research Report No. R-37. Lahore, Pakistan: InternationalIrrigation Management Institute.

Hamid, Naved; and Wouter Tims. 1990. Agricultural growth and economic development: The case of Pakistan. De-velopment Centre Technical Papers. Paris: Organization for Economic Cooperation and Development (OECD).

Haq, Asrarul; and Bagh Ali Shahid. 1997. Public, private, or participatory?: Reforming irrigation management in Pa-kistan. ICID Journal 46(1).

32

Hassan, Mehmoodul; Zafar Iqbal Mirza; and D. J. Bandaragoda. 1996. Mobilizing social organization volunteers: Aninitial methodological step towards establishing effective water users organizations. IIMI’s Pakistan National Pro-gram. Research Report No. R-9. Lahore, Pakistan: International Irrigation Management Institute.

Keller, A.; J. Keller; and D. Seckler. 1996. Integrated water resource systems: Theory and policy implications. Re-search Report 3. Colombo, Sri Lanka: International Irrigation Management Institute.

Kikuchi, Masao; Masako Fujita; and Yujiro Hayami. 1999. State, community and market in the deterioration of an irri-gation system in the Philippines. In Communities and markets in economic development, ed. M. Aoki and Y.Hayami. New York: Oxford University Press (forthcoming).

Meinzen-Dick, R. 1996. Groundwater markets in Pakistan: Participation and productivity. IFPRI Research Report No.105. Washington D.C., USA: International Food Policy Research Institute.

Meinzen-Dick, Ruth; Richard Reindinger; and Andrew Manzardo. 1995. Participation in irrigation. Environment De-partment Papers. Paper No. 003, Participation Series. Washington D.C.: The World Bank.

Memon, Y.; M. U. Hassan; and D. J. Bandaragoda. 1997. Socioeconomic baseline survey for three pilot distributariesin Sindh province. IIMI’s Pakistan National Program. Research Report No. R-36. Lahore, Pakistan: InternationalIrrigation Management Institute.

Merrey, D. J. 1993. Institutional context for managing irrigated agriculture. Paper presented at the DSE/IIMI StrategyWorkshop on Institutional Framework for Irrigation, 1-5 November 1993, Chiang Mai, Thailand. In Institutional frame-work for irrigation: Proceedings of a workshop, ed. C. Abernethy, 7–22. Colombo, Sri Lanka: International Irriga-tion Management Institute. (1996)

Ministry of Food and Agriculture, Government of Pakistan. 1988. Report of the National Commission on Agriculture.Islamabad, Pakistan.

Mirza, Ashfaq H. 1989. The dilemma of organizing water users associations in the Punjab. In Societal dynamics andirrigation systems, ed. Jamshed Tirmizi and Bradley Parlin. Islamabad, Pakistan: Pakistan Council of Researchin Water Resources.

Nasir, Ahmad. 1992. Decentralization: A study into forms, structures and development processes. Peshawar, Paki-stan: Pakistan Academy of Rural Development.

Oakerson, R. J. 1992. Analyzing the commons: A framework. In Making the commons work: Theory, practice and policy,ed. D. W. Bromley. San Francisco, California: Institute for Contemporary Studies.

Ostrom, E. 1992. Crafting institutions for self governing irrigation systems. San Francisco, California: Institute for Con-temporary Studies.

Ostrom, E. 1990. Governing the commons: The evolution of institutions for collective action. New York: CambridgeUniversity Press.

Restreppo, Carlos G.; D. J. Bandaragoda; and Pierre Strosser. 1994. Crop-based irrigation operations study in theNorth-West Frontier Province of Pakistan, Vol. I: Synthesis of findings and recommendations. IIMI’s Pakistan Na-tional Program. Research Report No. R-1. Lahore, Pakistan: International Irrigation Management Institute.

Runge, C. F. 1992. Common property and collective action in economic development. In Making the commons work:Theory, practice and policy, ed. D. W. Bromley. San Fransisco, California: Institute for Contemporary Studies.

Samad, M.; and Douglas Vermillion. 1999. An assessment of the impact of participatory irrigation management in SriLanka. Water Resources Development 15 (1/2): 219-240.

Samad, Madar; and Pieter Smidt. 1999. Management transfer in irrigation systems in Asia: A review of selected countryexperiences. Draft paper prepared for presentation at the International Commission on Irrigation and DrainageConference, Granada, Spain.

Scheumann, W. 1997. Managing salinization: Institutional analysis of public irrigation systems. Berlin: Springer.

33

Schrijvers, J. 1995. Participation and power: A transformative feminist research perspective. In Power and participa-tory development: Theory and practice, ed. Nici Nelson and Susan Wright. London: Intermediate Technical Pub-lications.

Skogerboe, G. V.; L. P. Poudyal; and K. B. Shrestha. 1993. M&O guidelines for turnover of irrigation systems to farm-ers. Water Resource Development 9(4): 369-386.

Turral, Hugh. 1995. Devolution of management in public irrigation systems: Cost shedding, empowerment and per-formance. Working Paper 20. UK: Overseas Development Institute.

Uphoff, N. 1986. Improving international irrigation management with farmer participation: Getting the process right.Boulder, Colorado: Westview Press.

Vander Velde, E. J.; and D. H. Murray-Rust. 1992. Impacts of physical managerial interventions on canal performancein Pakistan: A review of five years of field research studies. Paper presented at the Eighth Internal Program Re-view. Colombo, Sri Lanka: International Irrigation Management Institute.

Vermillion, Douglas L; and Garcés-Restrepo. 1998. Impacts of Colombia’s current irrigation management transfer pro-gram. Research Report 25. Colombo, Sri Lanka: International Water Management Institute.

Wade, R. 1988. The management of irrigation systems: How to evoke trust and avoid prisoners’ dilemma. World De-velopment 16 (4): 489-500.

Water Sector Investment Plan. 1990. Water sector investment plan (1990-2000). Lahore, Pakistan: Federal PlanningCell.

World Bank. 1993. Water resources management: A World Bank policy paper. Washington D.C.: The World Bank.

World Bank. 1994. Pakistan irrigation and drainage: Issues and options. Report No. 11884-Pak. Washington D.C.:The World Bank.

World Bank. 1996. Irrigation investment in Pakistan. OED Precis No. 24. Washington D.C.: Operation Evaluation De-partment, World Bank.

Zaman, Waheed uz. 1998. Self-help maintenance activities by the water users federation of Hakra 4-R Distributary.IIMI’s Pakistan National Program. Research Report No. R-44. Lahore, Pakistan: International Water ManagementInstitute.

Research Reports

23. Performance of Two Transferred Modules in the Lagunera Region: Water Relations.G. Levine, A. Cruz, D. Garcia, C. Garcés-Restrepo, and S. Johnson III, 1998.

24. Farmer Response to Rationed and Uncertain Irrigation Supplies. C. J. Perry and S.G. Narayanamurthy, 1998.

25. Impacts of Colombia’s Current Irrigation Management Transfer Program. Douglas L.Vermillion and Carlos Garcés-Restrepo, 1998.

26. Use of Historical Data as a Decision Support Tool in Watershed Management: ACase Study of the Upper Nilwala Basin in Sri Lanka. W. K. B. Elkaduwa and R.Sakthivadivel, 1998

27. Performance Evaluation of the Bhakra Irrigation System, India, Using AdvancedInformation Technologies. Wim Bastiaanssen and D. Molden, 1998.

28. Performance Evaluation of the Bhakra Irrigation System, India, Using Remote Sensingand GIS Techniques. R. Sakthivadivel, S. Thiruvengadachari, Upali Amerasinghe,W.G.M. Bastiaanssen, and David Molden, 1999.

29. Generic Typology for Irrigation Systems Operation. D. Renault, and G.G.A.Godaliyadda, 1999.

30. Mechanically Reclaiming Abandoned Saline Soils: A Numerical Evaluation. S. A.Prathapar and Asad S. Qureshi, 1999.

31. Gender Issues and Women’s Participation in Irrigated Agriculture: The Case of TwoPrivate Irrigation Canals in Carchi, Ecuador. Elena P. Bastidas, 1999.

32. Water Scarcity Variations within a Country: A Case Study of Sri Lanka. Upali A.Amarasinghe, Lal Mutuwatta, and R. Sakthivadivel, 1999.

33. Modernization Using the Structured System Design of the Bhadra Reservoir Project,India: An Intervention Analysis. R. Sakthivadivel, S. Thiruvengadachari, and UpaliA. Amarasinghe, 1999.

34. Assessment of Participatory Management of Irrigation Schemes in Sri Lanka: PartialReforms, Partial Benefits. M. Samad and Douglas Vermillion, 1999.

35. Modernizing Irrigation Operations: Spatially Differentiated Resource Allocations. D.Renault and I. W. Makin, 1999.

36. Institutional Change and Shared Management of Water Resources in Large CanalSystems: Results of an Action Research Program in Pakistan. D. J. Bandaragoda,1999.

Research Report

International Water Management Institute

INTERNATIONAL WATER MANAGEMENT INSTITUTEP O Box 2075, Colombo, Sri Lanka

Tel (94-1) 867404 • Fax (94-1) 866854 • E-mail iwmi@cgiar.orgInternet Home Page http: //www.cgiar.org/iwmi

ISSN 1026-0862

36

ISBN 92-9090-387-2

Institutional Change and SharedManagement of Water Resourcesin Large Canal Systems: Resultsof an Action Research Program inPakistan

D. J. Bandaragoda