Governing Maintenance Provision in Irrigation2001)_Governing...Ringler (Germany),Mark Rosegrant...

Governing Maintenance Provisionin Irrigation

A Guide to Institutionally Viable Maintenance Strategies

Universum Verlagsanstalt

Deutsche Gesellschaft fürTechnische Zusammenarbeit (GTZ) GmbHISBN 3-88085-531-5

Go

vern

ing

Mai

nten

ance

Pro

visi

on

in Ir

rig

atio

n27

3

Schriftenreihe der GTZ, Nr. 273

Governing Maintenance Provisionin Irrigation

Walter HuppertMark SvendsenDouglas L. Vermillion

Governing Maintenance Provision in Irrigation

A Guide to Institutionally Viable Maintenance Strategies

with contributions from Birgitta Wolff, Martin Burton, Paul van Hofwegen, Ruth Meinzen-Dick, Waltina Scheumann and Klaus Urban

Wiesbaden 2001

UniversumVerlagsanstalt GmbH KG

Published by: Deutsche Gesellschaft für

Technische Zusammenarbeit (GTZ) GmbH

Postfach 5180 · 65726 Eschborn

Federal Republic of Germany

Internet: http://www.gtz.de

Responsible: Walter Huppert, GTZ

Authors: Walter Huppert, Mark Svendsen, Douglas L. Vermillion

Die Deutsche Bibliothek – CIP Cataloguing-in-Publication-Data

A catalogue record for this publication is available from

Die Deutsche Bibliothek

ISBN 3-88085-531-5

to Ian Carruthers †

Preface

This Guide is about maintenance in irrigation.However it is notabout how to do maintenance. It is about how to providemaintenance. Manuals and guides about how to do maintenanceabound. In fact, one may assume that “How to” manuals onmaintenance written in various languages outnumber any otherkind of guide and manual in irrigation.In contrast to this abundancethere is – to the knowledge of the authors – not a single guide onmaintenance provision.

The reader may ask:what in fact is the difference? In this Guidewe perceive maintenance not only as a technical activity but at thesame time as a service delivery by somebody to somebody,in otherwords, as a provision.This may be difficult to comprehend at firstsight. However, Ostrom et al. (1993) observe that it may becomeclearer when we remind ourselves of our daily economicexchanges.Here,we distinguish clearly between goods and serviceswe produce ourselves in the household and others which areprovided to us. This comes down to a distinction betweenproduction and provision. Confusion and misunderstandings arecreated by the fact that this distinction between production andprovision becomes less selfevident in the public and non-commercial realm.Sometimes,the unit of government that providesa facility or a service is also the producer of that very facility orservice. However, frequently, a public agency may construct afacility,but rely on private agents to provide services like operationand/or maintenance of that facility.Or else,the agency may contractout all of these tasks.

In this Guide, we contend that major problems of maintenance– not only in irrigation – are due to the fact that maintenanceprovision is deficient or non-functional. Roles, rights andobligations between the providers and the beneficiaries ofmaintenance are unclear, agreements between these two sides arevague or non-existent, and mechanisms to enforce existingagreements are absent, as are the means to ensure the timelyprovision of necessary supporting services. In other words:deficient institutional arrangements for maintenance provisionare – to our opinion – at the roots of the maintenance crisis inirrigation and they are a “blank spot”in the maintenance discussionat the same time.

Preface

7

This is why this Guide promotes a “provision perspective” tomaintenance that is urgently needed in order to complement thenecessary but not sufficient “production perspective” that hasdominated the maintenance discussion so far.

Having defined the focus of the Guide, it becomes clear that thisis not a manual for engineers and technicians to help them improvemaintenance activities as they are implemented in the field. TheGuide hopes instead to help policy makers, managers, plannersand representatives of all the major stakeholders to develop a newperception of those “soft” institutional problems that need to beovercome when maintenance efforts are to be successful.

The provision perspective presented here can be applied wellbeyond the subject of irrigation maintenance. The concept andinstruments of this Guide can be used just as well for any otherservice provision in irrigation – from water delivery to drainage tosupporting services such as input supply, extension or dataprovision. Moreover, they are valid for fields of service provisionother than irrigation.

The Guide builds on the experiences and results of theMAINTAIN project, implemented by Deutsche Gesellschaft fürTechnische Zusammenarbeit (GTZ) and financed by the GermanMinistry for Economic Cooperation and Development (BMZ).

The concept and the recommendations are the result of anumber of case studies and thematic papers that have beenelaborated by various authors in the context of the MAINTAINproject of GTZ. These MAINTAIN papers are listed in the sectionon references and literature at the end of the volume.The ideal wayto profit from this Guide is therefore to use it in conjunction withthese documents, and the Guide contains numerous references tothese publications.The Guide,however,is comprehensive and self-contained and can also be used independently.

Preface

8

Acknowledgements

This Guide and the series of Case Studies and Thematic Papersit is based on, are the outcome of the “MAINTAIN project”implemented by Deutsche Gesellschaft für TechnischeZusammenarbeit (GTZ). The project was funded by the GermanMinistry for Economic Cooperation and Development (BMZ).Without the continuous support of Dr. Hans-Joachim de Haas andhis division these results could not have been achieved.

As far as the concept of maintenance provision is concerned,weowe much to two prominent authors and their works: Firstly toPhilipp Herder-Dorneich whose publications have spurred muchof the discussion on the side of the German contributors toMAINTAIN. Unfortunately, his book “Theorie der SozialenSteuerung” (“Theory of Social Governance”) is only available inGerman language. Secondly to Elinor Ostrom and her numerouspublications on irrigation institutions. The book “InstitutionalIncentives and Sustainable Development” published by hertogether with Larry Schroeder and Susan Wynne has providedessential foundations to the concepts and arguments of MAINTAIN.

Particular thanks are due to those members of the MAINTAINteam whose contributions have only been temporary butnonetheless essential:Artur Vallentin,Thomas Buhl-Böhnert,HeikeKühlwein-Neuhoff, Susanne Arlinghaus and Christian Hagen.

Apart from the authors and the MAINTAIN-team,a great numberof experts and discussion partners in various countries havecontributed ideas and information to single MAINTAIN papers, toworkshops, seminars and informal conversations that formed thebasis for this Guide. Particular thanks are due to (alphabeticalorder):Bilal Bashir (Jordan),David Cone (USA),Ashok Gulati (India),Mehmood Ul Hassan (Pakistan), Thilo Hatzius (Germany), SamJohnson III (USA), Wim H. Kloezen (the Netherlands), Gil Levine(USA), Doug Merrey (USA), Jamie Morrison (UK), Raymond Peters(India), Mike Paisner (USA), Jochen Renger (Germany), ClaudiaRingler (Germany),Mark Rosegrant (USA),John C.Skutsch (UnitedKingdom),Gert Soer (the Netherlands),Henri Tardieu (France) andPhilippus Wester (the Netherlands).

Special thanks go to the International Water ManagementInstitute (IWMI) and to the International Food Policy Institute(IFPRI). IWMI has cooperated with GTZ in contributing papers to

Acknowledgements

9

the MAINTAIN Series and supported workshop activities related tothe MAINTAIN concept. IFPRI has also contributed to theMAINTAIN papers. Moreover, IFPRI was generous enough to behost for one of the authors (Walter Huppert) in Washington,so thathe could work as a Visiting Research Fellow on the MAINTAINproject.

The MAINTAIN papers and this Guide would not have seen thelight in their present form without the assistance and thescrupulous editing work of Monica de Gregorio and Oliver Adam.

The authors accept sole responsibility for the content of thisGuide.

Acknowledgements

10

Table of Contents

Preface 7Acknowledgements 9Summary and Introduction 17

Part One: From Maintenance “Production” to Maintenance “Provision” 21

1. The Maintenance Problem 22

2. The Conceptual Framework 262.1 Maintenance in the context

of infrastructure service provision 282.2 Performance orientation

of maintenance provision 322.3 Inherent problems of maintenance provision 342.4 The disciplinary domains

of maintenance provision 382.5 Irrigation and maintenance

as multi-actor enterprises 40

3. Institutional Requirements for Maintenance Provision 433.1 The external institutional environment 443.2 Service arrangements 46

4. Developing Institutionally Viable Maintenance Strategies 514.1 “Strategic Institutional Positioning” 514.2 Analysis and improvement

of service arrangements 59

Contents

11

Part Two: Analysing and Improving Maintenance Provision 61

Using the MAINTAIN-Modules to develop institutionally viable maintenance strategies and service arrangements 63

Module 1: Identifying Service Objectives and Performance Standards 65

Module 2: Rapid Asset Appraisal (RAA) 82

Module 3: Rapid Assessment of Economic Incentives for Maintenance (REA) 100

Module 4: “Service Interaction Analysis” (SIA) 107

Module 5: Analysis of Property Rights and Authority Systems 119

Module 6: Analysing and Improving the Governance of Maintenance Provision 126

Module 7: Analysis of “Principal-Agent” problems 143

Module 8: Free Riding – A Central Governance Problem related to Maintenance 154

Module 9: Institutional Arrangements for Irrigation Financing 161

Module 10: Actor Specific “Incentive Profiles” 172

References and Literature 185

Authors’ addresses 190

Contents

12

Tables, Figures and Boxes

13


Table 1: Examples of incentive deficiencies for irrigation maintenance 24

Table 2: Comparison of the production perspective and the provision perspective for maintenance in irrigation 27

Table 3: Performance services of physical components of irrigation and drainage schemes 30

Table 4: Organizations, entities and groups receiving and providing services / supporting services from and to the Nienburg Association 42

Table M0: Applying the MAINTAIN Guide in differing contexts 64

Table M1-1: Relationship between level of investment, canal control, level of service and O&M requirements and costs 70

Table M1-2: Possible performance, condition and importance relationships 77

Table M1-3: Criteria for assessing level of service provision (of irrigation water supply) from farmers’ perspective 79

Table M1-4: Possible Irrigation Serviceability Matrix 80

Table M1-5: Gaps between levels of service 81

Table M2-1: Performance services of physical components of irrigation and drainage schemes 88

Table M2-2: Possible performance, condition and importance relationships 90

Table M2-3: Summary of steps for Rapid Asset Appraisal 92

Table M2-4: Example for proforma for asset survey data collection 94

Table M2-5: Example of standardised condition grading for a concrete bridge over a canal or drain 96

Table M2-6: Example of standardised performance grading for a canal head regulator 98

Table M2-7: Asset function index for determination of importance grading for irrigation and drainage infrastructure 99

Table M4-1: Example of a Hydro-Institutional Service Chain 111

Table M4-2: Range of services provided by German Maintenance Associations in general and the Nienburg Association in particular 114

Table M4-3: Range of services provided toGerman Maintenance Associations 115

Table M5-1: Authority system for Andhra Pradesh irrigation management reforms 121

Table M5-2: Categories of rights, duties and service functions 123

Table M6-1: Major governance modes and coordination mechanisms 127

Table M6-2: Institutional arrangements for maintenance provision 129

Table M7-1: Types of Principle-Agent problems and solutions in service relationships 149

Table M10-1: Examples of incentive deficiencies for irrigation maintenance 173

Table M10-2: Actor-specific Incentive Profile (ASIP) for maintenance service provision by CACG/France 182

Table M10-3: Actor-specific Incentive Profile (ASIP) for maintenance service provision in many Indian/Pakistani irrigation schemes 184

Figure 1: The major disciplinary domains related to the issue of maintenance 39

Figure 2: Institutional arrangements for service provision 44

Figure 3: Commercial irrigation service provision: integration of functions at the client side 47

Figure 4: Irrigation service provision by government entity to water users: splitting of functions at the client side 48

Figure 5: Strategic Institutional Positioning 52

Figure 6: Situation specific maintenance strategies in differentinstitutional environments 56

Figure M1-1: Maintenance causality chain 65

Figure M1-2: Different approaches to objective determination in different institutional contexts 75

Figure M3-1: Matrix for Rapid Economic Assessment of maintenance needs 106

Figure M4-1: Field of actors involved in maintenance service provision 112

Figure M4-2: Services provided by the Nienburg/Weser Maintenance Association: analysis of strengths andweaknesses 117

Figure M6-1: Key services and service relations in the “périmètres en concession” 132


14

Figure M6-2: Situation specific solutions to governing maintenance provision 141

Figure M9-1: Private service provision 170

Figure M9-2: Government subsidy to private service provision 170

Figure M9-3: Government subsidy and split arranging function 170

Figure M9-4: Financially autonomous government agency 171

Figure M9-5: Government administration 171

Box 1: Private and collective goods 35

Box 2: Sequence of steps to be followed when analysing and improving service arrangements for maintenance provision 60

Box M1-1: Objective determination for maintenance in a California irrigation district 72

Box M1-2: Objective determination for maintenance in Peruvian smallholder irrigation systems 73

Box M2-1: Essential steps in Asset Management 83

Box M6-1: Services, returns and coordination mechanisms – the example of Franchiseperimeters in Southern France 131

Box M6-2: Using external arbitersin maintenance management 138

Box M6-3: “Agreement to agree” arrangements in Andean irrigation schemes 139

Box M7-1: Facing Moral Hazard problems in maintenance provision – an example 145

Box M7-2: Facing a Hold-Up problem in maintenance provision – an example 146

Box M7-3: Facing problems of Adverse Selection – an example 147

Box M8-1: Free-riding behaviour in maintenance provision – a rationality trap 155

Box M9-1: Government subsidy to private service provision – the case of Southern France 165

Box M9-2: Government subsidy and jointly assumed arranging function 167

Box M9-3: Service provision by financially autonomous government agency 168

Box M10-1: Implementing ASIP – The case of the “Neste System” in southern France 180

Box M10-2: Implementing ASIP – The case of medium andlarge irrigation systems in Pakistan and India 181

Tables. Figures and Boxes

15

Summary and Introduction

17


The figures are shocking: According to UNEP, some 1.5 millionhectares of irrigated land are lost every year as a result of salinisationand waterlogging. The FAO estimates that, world-wide,approximately 30 million hectares of irrigated land are severelydamaged, and a further 60-80 million hectares partially damaged.Various studies emphasise the fact that the underlying reasons forthis are to be found in the operation and maintenance (O&M) ofirrigation systems.As the World Bank puts it in its “Review of WorldBank Experience in Irrigation”,“Poor quality of project design andplanning are big problems, but poor operation and maintenanceis a bigger one”… “O&M problems can be seen in the Bank’sfinancing of so many rehabilitation projects.Almost all of them,when scrutinised, turn out to be deferred maintenance projects”.

In an analysis of its experiences in 614 irrigation projects, theWorld Bank found that 43% of all project evaluations made noreference at all to operation and maintenance issues.The situationis even worse when maintenance is considered in isolation. In thefew statistical studies which do exist, operation and maintenanceare rarely dealt with separately. Generally the bulk of attention ispaid to operational deficiencies and only in exceptional cases doirrigation project evaluations address maintenance issues in anygreater depth.The aforementioned World Bank analysis describesthe situation laconically: “… audits rarely pay much attention topoor maintenance”.

Given this situation, one can only conclude that irrigation facesa “Maintenance Paradox”. On the one hand poor maintenance isclearly the origin of many of the most serious problems faced bythe irrigation sector. On the other hand, maintenance seems to bea sort of “non-issue”. What are the reasons for this paradox, thisstriking discrepancy between the acknowledged importance ofmaintenance and the lack of attention it is given in irrigationpractice?

In this Guide we contend that a one-sided perception ofmaintenance is the major obstacle that must be overcome in orderto devise solutions to this paradox.The maintenance discussion stillconcentrates nearly exclusively on questions of “how to do”

maintenance and how to finance the needed activities. This“production perspective” normally is not concerned with theinvolved actors, their interactions, and the laws, rules, rights andformal or informal contracts that govern the relationships betweenthese actors. In other words: the existing and the necessaryinstitutional conditions for maintenance provision are not a subjectof consideration. No wonder then, that major obstacles tocoordination and motivation of the stakeholders in irrigationmaintenance can neither be detected nor overcome.

This is why we are not concerned here with the waymaintenance activities are implemented, i.e. with the waymaintenance is “produced”. This is the topic of myriad technicaland managerial books and manuals. Instead, our concern here iswith the question who makes this “maintenance product”availableto whom,how this provision is organized,and the incentives for itsprovision.

By the same token, the Guide is not concerned with narroweconomic and financial issues of water pricing and tariff setting inirrigation. These issues, important as they are, are dealt with atlength in numerous books and articles. However, the Guide doesconfront the often neglected problem of finance provision,i.e.thesource of the financing and the terms and conditions which governits supply.Financing is viewed as part of a service exchange,whereit constitutes the “return” for a service that has been or is to beprovided.The institutional arrangements required to make that partof the exchange relationship function are at the center of ourinterest.

In sum, the Guide intends to aid policy makers, managers,planners and representatives of all the major stakeholders todevelop an understanding of the “provision perspective” tomaintenance (and to other services) in irrigation and of theinstitutional issues involved.Given this focus and target group, it isevident that we cannot present recipes. Individual contexts andmaintenance problem situations are extremly diverse and subjectto many contingencies. Hence there cannot be “cook bookrecommendations”for how to cope with institutional deficiencies.However, the Guide does present different basic strategies for theimprovement of maintenance provision in different institutionalcontexts. In doing this, we reject the implicit assumption of


18

technically-oriented maintenance approaches that there is “onebest way” to organize maintenance provision. Based on these“situation specific” strategies, a number of concepts and practicalinstruments can be applied that are presented and discussed in away that allows their flexible and independent use.

The Guide tries to strike a balance between a theoretical conceptpaper and a practical manual.

In Part One, after briefly taking stock of the maintenanceproblem in irrigation (Chapter 1), we present a concept thatexplains the essentials of the perception of maintenance as aprovision (Chapter 2). We define “infrastructure services” inirrigation and determine the meaning of maintenance provision inrelation to such services. We explain the importance of aperformance orientation for maintenance provision and presentsome definitions of maintenance that underline this importance.After discussing some inherent problems of maintenanceprovision we expose a perception of water delivery andmaintenance as multi-actor service systems.Such a way of thinkingimplies that water delivery and maintenance will only functioneffectively and efficiently if the relationships between the variousactors are well-functioning.This requires good coordination but atthe same time sufficient motivation of the involved actors.Considerations like these direct attention to the institutionalrequirements for service provision (Chapter 3).We concentrate ontwo dimensions that have to be dealt with in any attempt to developor improve institutional arrangements: the institutionalframework and the institutional arrangements for a particularservice provision, in other words, the “service arrangement”.Based on these two dimensions we present a methodology –“Strategic Institutional Positioning” – that allows differentiatingstrategies for maintenance provision in different institutionalcontexts (Chapter 4).Finally,a rough sequence of steps is proposedthat may guide efforts to analyse and improve service arrangementsfor maintenance provision.

Part Two of the book makes available to the reader a range of tenpractical Modules.These Modules elaborate more in detail each ofthe steps proposed before. Thus, they touch upon topics likeobjective setting for maintenance provision, they presentmethods for rapid asset appraisal and rapid assessment ofeconomic incentives for maintenance and they deal in detail with


19

various topics relevant when analysing and improving institutionalarrangements for maintenance.Among these,the discussions on theissue of “governance” of maintenance provision are of centralimportance. Modules on “institutional arrangements formaintenance financing” and on “actor specific incentive profiles”conclude the volume.

The Guide is based on and complemented by a series ofMAINTAIN Case Studies and Thematic Papers. The reader who islooking for detailed references and literature should consult thesedocuments. When using the Modules in the second part of thisbook, indication is given as to the MAINTAIN Papers that are mosthelpful to consult additionally.

However,even on its own,the Guide can be used in the followingways:� The Guide may be used as an introductory text on the “provision

perspective” to maintenance;� it may be used as a guide to design situation specific maintenance

strategies for irrigation maintenance;� it may be used in a flexible way to provide particular instruments

that help cope with specific problems of maintenance provision.


20

21

Part One

From Maintenance “Production” to Maintenance “Provision”

1. The Maintenance Problem

The United Nations predicts that the world’spopulation will increase by about 3 billion between2000 and 2050. The FAO has argued that 60 % of theadditional food required by this growing populationwill have to be produced on irrigated land. In manycountries the yields for food crops have leveled off overthe past decade.Moreover the amount of good new landwhich can be devoted to growing crops is extremelylimited.However, there remains considerable potentialto improve the yield of food crops per unit of irrigationwater delivered. Hence, much of the increase in foodproduction in the future will need to come fromimprovements in the operation and maintenance ofexisting irrigation systems.

Despite this powerful need to improve irrigationmanagement, the performance of irrigation systemshas, in general, been disappointing (especially those inthe public sector). Large sums have been spent forconstruction, rehabilitation and modernization, whilecomparatively small amounts have been spent foroperations and maintenance. Maintenance tends to bedeferred to the future, in anticipation of externalfinancing for rehabilitation. Deterioration of publicirrigation systems in developing countries is rapid andalmost universal, resulting in loss of production andfrequent rehabilitation.

A World Bank study found that approximately 66 %of funds invested annually for irrigation developmentare for “premature rehabilitation”, which was madenecessary by deferred maintenance. In manydeveloping countries with over-staffed bureaucracies,the large majority of funds available for maintenance isused to cover personnel costs. Little is left for actualmaintenance. Corruption and undue influence frompartisan interests lead to misallocation of funds, faultyconstruction and other inefficiencies. Farmersdissatisfied with irrigation services are unwilling to pay

22


irrigation service fees, which further inhibitsmobilization of sufficient funds for maintenance.

The pattern of “build-neglect-rebuild” and poormaintenance management have widespread andserious consequences. These include the shrinking ofirrigation service areas, inefficient and inequitabledistribution of water, loss of capacity to measure andcontrol water, and waterlogging of otherwiseproductive land. These problems result in loss ofagricultural productivity, declines in farm income,inability to collect water charges from farmers, andsubstantial debt burdens on governments which mustrepay loans for premature and repeating rehabilitationprojects.

Since inadequate maintenance of irrigation systemshas such serious consequences, the relative lack ofattention by governments and internationaldevelopment agencies to the maintenance problem isremarkable. It is this lack of attention to maintenancethat we refer to as the maintenance paradox. Theparadox is: If the maintenance problem is so serious,why is there so little attention given to solving it?

Disincentives for adequate maintenance of irrigationsystems effect senior government officials, irrigationagency staff, farmers and international developmentagencies. These disincentives are especially strong incentrally-managed irrigation bureaucracies where thegovernment still acts as the primary provider, rights-holder and payer of irrigation services.

Table 1 displays common disincentives towardirrigation maintenance affecting key stakeholders in aconventional top-down administrative setting. Thesedisincentives are especially pronounced in such asetting but they are not unique to it.Disincentives leavefarmers, governments and international agencieswithout sufficient motivation to invest in maintenanceat levels which would ensure the functionalsustainability of irrigation systems.

This Guide adopts the premise that the general lackof commitment toward solving the problem of

23


24


Table 1. Examples of incentive deficiencies for irrigation maintenance

Stakeholder Incentive Deficiencies

Senior � Low political benefits, high opportunity costsGovernment � Low, delayed visibility of benefits of maintenanceOfficials � Low budget priority. Rehabilitation projects create

political support

Irrigation Agency � Budget allocations unrelated to fee collection ratesManagement � Internal political benefits of maximizing employment

rolls

� Accountability to internal hierarchy simpler thanaccountability to water users

Operational Staff � Maintenance lacks professional appealof Irrigation � Deterioration rewarded by rehabilitation projectsAgencies � Accountability to internal hierarchy instead of

accounting to water users

� Internal accountability mechanisms do not rewardgood maintenance

Water Users � Irrigation infrastructure seen as government propertyand responsibility of government to maintain

� No relation between payment of water fees andquantity or quality of maintenance

� No clear water rights

� Not involved in priority setting for maintenance works

Foreign Donors � Difficulty in monitoring the use of resources formaintenance

� Difficulty in monitoring the benefits of effectivemaintenance

� Pressures to perpetuate the financing of capitalintensive projects, such as rehabilitation,modernization and expansion

� Reluctance to fund recurrent costs

inadequate maintenance of irrigation systems has itsroots in a one-sided and partial understanding of whatmaintenance is about. This perception focussesexclusively on the questions of how to do maintenanceand how to pay for such maintenance activities. In thefollowing we refer to this perception as the productionperspective.

However, maintenance has two sides, like the twofaces of a coin.What we need to consider additionally,in order to approach solutions to the above-statedmaintenance paradox and the related incentivedeficiencies is an understanding of maintenance as aprovision. Major problems of maintenance, in thisperspective, are due to the fact that maintenanceprovision is not well structured or does not functionaltogether. More seriously, in most cases key questionsof maintenance provision are not discussed at all.Whoprovides maintenance to whom? Who are otherstakeholders involved in the provision process andwhat functions do they assume or what supportservices do they provide? What are the mechanisms thatgovern the relationships and exchanges between thestakeholders in these provision processes? On the basisof what kind of agreements, rights, contracts orcommon practices does the provision of maintenanceand of the supporting services occur? Do thesearrangements provide incentives to engage in therelationship? How can non-compliance with provisionagreements be enforced? These and other questionsrelating to institutional arrangements formaintenance are at the center of a perception we referto as the provision perspective.

We contend that a comprehensive treatment of thetopic of maintenance requires consideration of boththe provision and the production of maintenanceservices.

25


2. The Conceptual Framework

As mentioned above, this Guide builds on aconceptual framework – the provision perspective –that differs substantially from the dominantmaintenance paradigm. The key elements of thisconcept are the following (see also Table 2).� As stated before, maintenance is defined here as a

provision and not simply as a technical task. Moreprecisely, maintenance is looked upon as a serviceclosely linked to the major infrastructure services inirrigation: the provision of infrastructure and waterdelivery.

� This Guide raises awareness of the fact thatmaintenance services are prone to particularinherent problems that need to be taken intoaccount.Amongst others,maintenance services oftenhave the features of collective goods or club goods.Moreover, they are so-called “future goods” in manycases – i.e.the benefits of maintenance often accruein a distant future.And maintenance provision oftenis highly intransparent, i.e. it is difficult in manysituations to check whether or not maintenance hasactually been done. Characteristics like these mayinduce particular consequences and requireparticular approaches to deal with them.

� In this Guide we contend that maintenanceproblems are related to three distinct domains thatare highly interrelated:– The technical domain– The economic/financial domain– The institutional/organizational domainEmphasis in this Guide is given to theinstitutional domain which has been widelyneglected so far when dealing with maintenanceissues.

� Irrigation in general and maintenance in particularare perceived as multi-actor enterprises. Serviceprovision involves different roles: the roles ofprovider,arranger,payer,consumer and often also the

26


27


Table 2: Comparison of the production perspective and theprovision perspective for maintenance in irrigation.

Production Perspective Provision Perspective Reference inthis Guide

Maintenance defined as Maintenance defined as Para 2.1a technical task a service provision

Maintenance as a Maintenance as a secondary Para 2.1complementary task to service to the primary services system operation (O&M) of “infrastructure provision” (PI)

and “water delivery”

No attention to special Deals with consequences Para 2.3“goods character” of special “goods character” Module 8of maintenance of maintenance (collective

or club good; future good; good with low transparency)

Focus on the technical Focus on the institutional Para 2.3 andand the financial domain domain chapters 3

and 4Modules 5 to 10

No attention to involved Special focus on involved Para 2.4stakeholders stakeholders and their roles, Modules 4 and 5

rights and functions

No attention to questions Central focus on the Chapter 3of coordination and governance of service Modules 6, 7 motivation of stakeholders relationships and hence on and 9

coordination and motivation of stakeholders

Often (but not necessarily) Performance orientation Para 2.2low attention to emphasized through Module 1performance orientation definition of “level of

maintenance service provision”

No situational Emphasis on situational Chapter 4differentiation of differentiation of strategies Implicit inmaintenance strategies for maintenance provision all Modules

28


roles of arbiter, auditor and regulator. In most casesthese roles are assumed by different organizations,groups or individuals. A core requirement forsuccessful maintenance service delivery is thecoordination and motivation of these multipleactors towards a common performance goal.

� To achieve such coordination and motivation meansto achieve accountability. The Guide defines anddiscusses the two important dimensions of account-ability:– the service arrangement and– the institutional frameworkStrategies for the improvement of maintenance willlargely depend on the strength of the servicearrangement that is in place and of thesupportiveness of the institutional framework.

� Given the service perception of maintenancepromoted by this Guide,a performance orientationof maintenance – a defined “level of maintenanceservice provision” – needs to be introduced. Arealistic level of this kind will necessarily be afunction of the technical and economic constraintsthat impinge on the irrigation system.However,hereagain, the definition of this level will also dependgreatly on the quality of the service arrangement andthe institutional framework conditions.To define andachieve an optimal level of maintenance provisionin diverse situations set by these constraints is hencea challenge to which this Guide hopes to make acontribution.

2.1 Maintenance in the context of infrastructureservice provision

In order to understand service provision in irrigationand the role of maintenance, it is useful to look moreclosely at what is meant by infrastructure services.Infrastructure services are those services that can beprovided by means of a given piece of infrastructure.

What are

“infrastructure

services”?

Unbundling sets of infrastructure services reveals thefollowing:� The most general infrastructure service consists of

making infrastructure available for use,the provision-of-infrastructure (PI) service. For example, a road ismade available by government for use by car traffic;a house is rented out by the owner to be used by atenant; a car is rented out to a temporary user.In some cases such primary PI-services may besubdivided into subsets of PI-services. For example,one PI service provider might construct a railway,while another makes available rolling stock. PIservices in road transport might consist of provisionof the roadways themselves, provision of trucks andprovision of goods containers.

� Subsequent to the PI service, some infrastructureservices consist of the performance of certainfunctions by means of the respective infrastructure,e.g.the delivery of certain material and immaterialgoods.Examples of such “performance services”(PS)are the conveyance and delivery of drinking water bymeans of a buried pipe system, the conveyance anddelivery of electrical current through a net oftransmission line, or the conveyance and delivery ofirrigation water by means of a network of irrigationcanals.Also,the function of a pump station “to lift water fromlevel A to level B” can be perceived as such aperformance service. The service of the puretransport and distribution of a good to the points ofdelivery may be enriched by having the servicedeliver the good in a certain quantity and quality,at acertain time and at a particular point.Table 3 lists theperformance services provided by various physicalcomponents of irrigation and drainage schemes.

� These two primary services, i.e. the provision-of-infrastructure service (PI) and the performanceservice (PS) can only be provided if the providersthemselves receive some indispensable (internally orexternally provided) secondary services. In the case

29

2.1 Maintenance in the context of infrastructure service provision

30


Table 3: Performance services of physical components ofirrigation and drainage schemes (Burton 2000 in MAINTAINThematic Paper No. 8)

Component Levels Performance service

Canals Primary To convey waterSecondaryTertiaryQuaternary

Drains Primary To remove water from the fieldSecondaryOn-farm

River weir Main canal To divert and control irrigation supplies

Headworks Main canal intake To take in water to the main canal. This may be agroup of structures, including a river weir, headregulator, settling basin, and measuring structure,or one structure such as a pump station.

Pump station Main canal To lift water to command level forMain drain irrigation. To remove water from drainage

channels which are below river level

Settling basin Main intake canal To settle out sediment

Cross regulator Primary and To raise and maintain water surfacesecondary canals at design elevation

Head regulator Primary, secondary To regulate discharge and tertiary canals entering a canal

Measuring Primary, secondary To measure dischargestructure and tertiary canals for operational purposes

Aqueduct All levels of canal To pass canal over an obstruction (another canal,a drainage channel, etc)

Culvert All levels of canal To pass canal or drain under anor drain obstruction (road, drainage channel, etc)

Drop structure All levels of canal To “drop” the canal or drain bed levelor drain in a safe manner. Used to slacken canal or drain

slopes on steep land

Escape structure All levels of canals To release water from a canal into the drainagenetwork in the event of oversupply or under-utilisation.

Syphon underpass All levels of canals To pass the canal below an obstruction such as aroad or drainage channel.

Distribution box Quaternary canal To distribute water between quaternary channels

Night storage Main canal or To store irrigation water during the night forreservoir on-farm release during the day. Main canals can thus

operate 24 hours/day whilst lower order canalscan be operated during the daytime.

Tubewell On-farm To abstract groundwater for irrigation. Often usedin conjunction with surface water system

Bridges Road bridges To allow human and animal traffic over the canalFoot bridges or drain

Roads Inspection roads To gain access to the irrigation system andAccess roads villages. For inspection and maintenance

of the PI services these are services of maintenance,rehabilitation and modernization of the infra-structure. In the case of the PS-services these areservices of system operation and accompanyingmaintenance (“O&M”).

� Generally,a number of other services are required toenable and facilitate the primary PI and PS servicesdescribed above.These include engineering design,contracting, billing, budgeting, financialmanagement, coordinating with other actors, publicrelations and so on. We term these supportingservices. We distinguish them from primary andsecondary PI and PS services in terms of thedirectness of their relationship with the process ofarranging and providing the irrigation water toclients. While measurement of water flows is anintegral part of delivering irrigation service,preparing bills for the service is only indirectlyrelated and is considered a supporting service.External organizations and groups are sometimescalled upon to provide these supporting services,though they can be supplied internally as well.

In such a context, maintenance can be perceived asa service to the supplier of the infrastructure, sincemaintenance is required to keep the infrastructure ingood condition so that it can be used to deliver a serviceto clients.Alternatively,maintenance may be perceivedas a service to the user of the infrastructure, enablingthe user to deliver a certain good by employing thisinfrastructure. Whether and to what extentmaintenance is a service to the PI agent or to the PSproviders is determined by the property rightsassociated with the infrastructure and by the terms ofthe agreement between these two parties. This opensthe possibility that maintenance obligations may besplit. For example, the supplier of the PI service mayremain responsible for maintenance related to the long-term preservation of the infrastructure asset,while theprovider of the PS service may be responsible for

31

2.1 Maintenance in the context of infrastructure service provision

maintenance related to day-to-day operation of theinfrastructure facility.

2.2 Performance orientation of maintenanceprovision

Vague or non-existing target levels for irrigationservices are among the major causes for incentivedeficiencies. If such “levels of service” have not beendefined, there will be no benchmark against which tojudge the efforts and contributions of the variousactors. But how to create incentives for a high qualityservice provision or for improvements of the actualprovision if these performance levels have not beendefined? How to judge maintenance provision if notarget level for such a provision has been specifiedbeforehand? Thinking in such terms, it is quiteremarkable that Burton states in MAINTAIN ThematicPaper No. 8, that “to the author’s knowledge little, ifany, work has been done on assessing farmers’desiredlevel of service in smallholder irrigation schemes indeveloping countries”.

To define an agreed upon target level of serviceprovision is essential for other reasons as well.Differentactors may have different aspirations and diverginglevels of expectations may become the source ofconflicts and disincentives. Examples abound inirrigation where ambitious scheme performance levelshave been formulated in the planning phase but where

32


In this Guide, we will return again and again to the character of maintenanceas a service provision. In Module 4 of Part Two we present an instrument thatallows identification of the services and supporting services that are providedin a complex network of involved actors. In chapter 2.3, we describe anumber of inherent problems related to the service character of maintenance.In Modules 5 to 8 we present ways to overcome some of these problems.

farmers failed to contribute to the achievement of suchtarget levels.

This is why, in this MAINTAIN-Guide, we advocateefforts to define service objectives and performancestandards as essential for improvements in maintenanceprovision. Doing this we point to recent definitions ofmaintenance such as those mentioned below. Thesedefinitions include a reference to system performanceand often specify a particular level of performance asthe criterion for successful maintenance. Particularlythe first two definitions place very clear emphasis onsystem functionality – its output – rather than onproviding a given input level for the maintenanceprocess.

Performance orientation is an essential element ofwhat we mean by “service orientation”: a change froma supply driven,input-oriented perspective to a demanddriven, output and performance oriented view ofmaintenance efforts.Using the insights we have gainedbefore with respect to the character of maintenance asa service provision, we base the discussions of thisGuide on the following MAINTAIN definition ofmaintenance.

33

2.2 Performance orientation of maintenance provision

Maintenance is...

... the upkeep of facilities with the goal of efficient operation, minimumbreakdowns, good appearance, reasonable costs, extended useful life, andsafety – Krause and Temple (1988)

... a management response to the deterioration of the physical condition ofirrigation systems that threatens to make it impossible to achieve operationaltargets – Karunasena (1993)

... any action required to either return an irrigation system to or keep it at adesired performance level – Thoreson et al. (1997)

34


2.3 Inherent problems of maintenance provision

When we perceive maintenance as a service we needto be aware that service provision is prone to aparticular set of problems.

As mentioned before, the particular problems ofmaintenance services emerge when we look at therelation among the exchange partners of such services.

The roots of such problems go back to the facts that� Maintenance services often have the features of

“collective goods”or “club goods” (see Box 1)� Maintenance services generally have the features of

“future goods”� Maintenance services in many cases are highly

intransparent services.

All of these facts contribute to one central problemin maintenance service provision: it may be verydifficult for the clients or customers of the service tosufficiently influence the provider to ensure thatprovision corresponds to clients’ needs. We call thiscentral problem the problem of “feedback

What are inherent

problems of

maintenance service

provision?

The MAINTAIN definition of maintenance

Maintenance is …

... both a technical activity and a service provision aimed at keeping irrigationinfrastructure at a desired performance capacity or to restoring it to aparticular capacity. It is a service supplied to the providers of theinfrastructure and/or to those who deliver certain goods by means of thisinfrastructure.

In Module 1 we discuss in detail approaches to objective determination forirrigation maintenance. In this Module, we also deal with the question of howto modify objective determination under different institutional conditions.

35


deficiencies”.Such feedback deficiencies are among theprime causes for incentive problems related tomaintenance (not only in irrigation).

Maintenance as a collective good or club good

In many cases a maintenance service is a so-called“collective good”, more precisely, a “club good”. Thismeans,that the service is not provided for an individualbut for a group. For collective goods, sometimes somepeople in the group that do not pay for the servicecannot be excluded from benefiting from it.

What are

characteristic

problems of services?

Box 1 – Private and Collective Goods

Private goods are goods (or services) we would normally acquire by purchase(e.g. an item of clothing, a hair cut). These are goods to which the so-called“exclusion principle” applies, i.e. those individuals who have not paid forthem can be excluded from their consumption. Private goods are also goodsto which the “competition principle” applies. This means that theconsumption of a unit of this good by a consumer reduces the availability ofthe good to other consumers by a certain degree.

Collective goods are goods to which the two above-mentioned principlessometimes do not apply. Where the non-excludability relates to open accessto the public, e.g. the use of goods such as public safety or an anti-air-pollution measure, then the term public goods is applied. In such cases, amember of the public who has not paid – i.e. in this context not paid any tax– cannot be excluded from consumption of the goods. The competitionprinciple is also not applicable, in that the “consumption” of public safety bya “consumer” does not have any adverse effect on correspondingconsumption by other beneficiaries.

Where the non-excludability of certain collective goods relates to a certaingroup of consumers – e.g. motorists on a road where a toll is levied – theterm club goods is applied. This denotes goods which potential consumersnot belonging to the club (e.g. of those who have not paid the toll) can beprevented from enjoying. Within this group, the competition principle is of nosignificance. The use of the good in question by a member of the group in noway constrains the use of the good by the others.1

1 This example clearly illustrates that there is rarely such a thing as “purely”public goods such as public safety. In the case of motorway use, froma certain number of users upwards the traffic jamx caused can indeedbring the competition principle into play.

The problems associated with such goods make itdifficult for providers and clients to “control” theirprovision. With goods of this type there is no market,as non-payers – so-called “free riders”– are also able toconsume the good in question. This also means that noprice can be formed in the strict sense, i.e. there is no“equilibrium price” to harmonise supply and demand.(Notwithstanding the fact that fees can nevertheless beset). From the lack of price formation it follows thatindividual clients also lack a key means of influencingthe supply. Consequently, also a “feedback deficiency”emerges between the service provider and the client:The provider cannot exclude non-paying users, and inturn the individual clients have no direct means ofinfluencing the quantity or quality of service provision.This normally leads to a serious breakdown ofincentives for further service provision. Evencustomers who are willing to pay for the service willcease to do so when they become aware that extensivefree-riding exists. With increasing free-riding thefinancial basis for the provision of the service will erodeand provision will eventually stop.

Maintenance as a “future good”

The maintenance service which is exchangedbetween a supply and a demand side has a characteristicwhich has important impacts on incentive creation: itis a “future good”. Future goods are goods or services,the benefits of which do not emerge until some pointin the future, but which have to be paid for in thepresent.A well-known problem with future goods is thefact that they are subject to the law of “undervaluationof future goods” (a law well known in the insurancebusiness).This law states that many consumers have a

36


The free-riding problem is dealt with in Module 8 and some strategies areprovided to help overcome such problems.

37


tendency to overly discount future needs and hence arenot willing to make the full necessary outlay in thepresent for goods which they cannot use until thefuture. Excessive discounting drastically reducesincentives for maintenance. Such a tendency will beparticularly pronounced with poor farmers indeveloping countries.

Maintenance as an intransparent service

The service features of maintenance entail a furtherpotential problem, the frequent intransparency ofservices. Intransparency means that, as a non-professional,the customer/client is unable to fully judgethe value of the service being provided by the provider,who is an expert. Water users, for example, seldompossess detailed engineering knowledge and hence, incases where they arrange the provision of maintenanceservices, it will be difficult for them to monitor andevaluate the more sophisticated maintenance works tobe done.

This lack of transparency inherent in someintransparent services automatically introduces apronounced “feedback deficiency” into the servicerelationship. Particular institutional arrangements areneeded to solve problems of this kind and to preventincentive problems having their roots in feelings of“loss of control”on the part of those who demand andbenefit from such maintenance services.

When looking at the economic incentives for maintenance, this discountingtendency and the resulting perception that different actors have of thepresent value of benefits needs to be taken into account. Module 3 of thisGuide introduces an approach for a rapid economic assessment that includessuch considerations.

Module 6 of this Guide introduces approaches that help to deal withfeedback deficiencies of this kind.

2.4 The disciplinary domains of maintenanceprovision

When thinking about maintenance problems,peopletend to relate these problems to the technicalinfrastructure: Silted-up irrigation ditches, embank-ments covered with weeds,rusted and warped sluices,sections of ditches with slope failure, undercut andbroken-off wing walls, weirs and drop structures, etc.are common symptoms of maintenance problems.These symptoms make it tempting to mistake effects forcauses, and to look exclusively for technical solutionsto maintenance problems. Since damage of this kind isoften not due merely to technical problems, this maysimply initiate a repeating cycle of technicalrehabilitations.

In this Guide, we perceive maintenance in thecontext of three major domains fitting within theoverarching framework conditions:

� The technical/physical domain (e.g., technology,design,construction,physical inputs)

� The economic/financial domain (e.g., costs,benefits, financing)

� The institutional/organizational domain (e.g.stakeholders, agencies, accountability, governance)

All of these domains need to be considered whendealing with maintenance in a comprehensive way.Asnoted in Table 2, the production perspective onmaintenance puts the major focus on the technical andfinancial domains.In contrast,the provision perspectiveconcentrates mainly on the institutional domain. Thisconcept can be visualized by a triangle embedded in anellipse that represents the overall frameworkconditions, as shown in Fig.1.

38


A detailed discussion of the technical domain is not the focus of this Guide.However, even in situations where institutional problems appear to bepredominant, the technical manifestations of institutional deficiencies need tobe assessed and documented.

Fig. 1: The major disciplinary domains related to the issue of

maintenance

An in-depth discussion of the economic domain inirrigation maintenance will have to examine not onlythe costs of maintenance, and the resulting level ofirrigation service,but also the benefit stream generatedby the service provided and the impact that differentlevels of maintenance have on this benefit stream.

This is a discussion which is neither attempted in thisGuide nor will it be a realistic undertaking in mostpractical irrigation situations.The complex cause-effectrelationships, discussed more in detail in Module 3 ofPart Two,are difficult to establish and the necessary dataare not available under normal circumstances. Still,

39

2.4 The disciplinary domains of maintenance provision

To satisfy this need, Module 2 in Part Two of the Guide presents a procedurefor “Rapid Asset Appraisal”. For more detailed information on Asset Appraisaland Asset Management techniques reference can be made to MAINTAINPaper No. 8.

incentives for stakeholders, especially for water users,to engage in maintenance are predominantly of aneconomic nature. Hence some economic “yardstick” isneeded when devising a detailed maintenance strategy.

The reader may also refer to a recent study by HRWallingford (Skutsch 1998),which highlights the issuesinvolved in the economic domain,without resorting toa fully detailed economic analysis.

2.5 Irrigation and maintenance as multi-actorenterprises

The so-called “administrative paradigm”has been thedominant approach toward irrigation development inmany developing countries since the colonial era.Medium and large irrigation schemes built bygovernment engineers and contractors were, afterconstruction,operated and maintained by governmentstaff who were provided O&M manuals and directedand financed from above. Administrative procedurescontrolled with little if any formal participation byfarmers.

The social landscape in developing countries haschanged greatly since the demise of colonialism. Thetop-down administrative paradigm has persisted inmany countries,at least superficially,despite increasingdemocratization, economic liberalization and thecommercialization of agriculture, which render itincreasingly obsolete.Today,irrigation schemes all overthe world generally consist of multiple stakeholderswho have contending interests and divergentperspectives.Hydraulic networks of irrigation schemestend to cut across socio-economic categories and local

40


To solve this dilemma, the Guide offers the concept of a “Rapid EconomicAssessment” of maintenance needs (already mentioned above) in Module 3.

government boundaries.Differences between head andtail ends of canals in water delivery service andmaintenance requirements generate social differencesand tensions.

As we have seen in section 2.1 and will furtherdiscuss in section 3.2 and Module 4,a number of distinctkey service roles or functions are directly involved inirrigation management services.These are to:� provide the primary irrigation services, i.e.

– provide the hydraulic infrastructure,– provide the water delivery service (water

capture,water conveyance,water distribution,water allocation etc.);

� provide the secondary services, i.e. those servicesthat are integral to the primary services (operationand maintenance);

� provide the supporting services (informationprovision,coordination, representation,etc.);

� arrange the primary, secondary and supportingservices (i.e. select providers,define terms ofreference,conclude agreements or contracts,monitor provision etc.);

� use the services; and� pay for the services.

These roles may be performed by different kinds ofentities, such as government agencies, water usersassociations, or contracting companies. Even if werestrict the discussion to the secondary service ofirrigation maintenance provision, there is normally alarge number of involved stakeholders.As an example,Table 4 reveals how vast the number of interrelatedactors can be in the field of maintenance. It lists all thedifferent organizations,entities and groups with whomthe Nienburg/Weser Maintenance Association inGermany has working relationships in order toaccomplish its purposes.

41

2.5 Irrigation and maintenance as multi-actor enterprises

42


Table 4 : Organizations, entities and groups receiving andproviding services / supporting services from and to theNienburg Maintenance Association in Germany.

(Source: Huppert and Urban, 1998)

1. Nienburg District Association (Umbrella organization)2. Lower District Water Authority3. Lower District Conservation Authority4. Upper District Water Authority5. Upper District Conservation Authority6. Independent Conservation Associations7. “29” Associations8. Farmers’ Association9. Members10. Obstructors (farmers objecting to rights of way)11. Contractors12. Own engineering offices13. External engineering offices14. Consultants15. Other interested parties16. Agriculture authorities17. Water management authorities18. Conservation authorities19. Banks20. Standards authority21. Subsidizing agencies22. Court of law23. Public prosecutors24. Neighborhood associations25. Fishery organizations26. Holders of water rights27. Communities28. Town and country planning authorities29. Raw material extraction companies30. Forestry authorities31. National and regional bodies32. Social environment

3. Institutional Requirements forMaintenance Provision

If we accept that irrigation maintenance is a serviceinvolving multiple actors with different interests andconstraints, then there are three key challenges forensuring effective service delivery.These are:

� To design the service provision process so thatinstitutional arrangements are compatible with theexisting institutional framework conditions.(To devise market-based arrangements for serviceexchanges,to take an example,in an institutional set-up without clear laws or traditions of property rightsand without strong and independent judiciarybodies that can enforce such laws, will be bound tofailure).

� To develop or improve institutional arrangementssuch that they bring about effective coordinationamong the involved actors.

� To design institutional arrangements such that theyprovide incentives which ensure motivation for allactors in a service arrangement to be accountable toone another for provision of the agreed service.

The terms “institutions” and “institutionalarrangements”in this Guide are used to indicate formaland informal rules and mechanisms (of a regulative,normative and cognitive nature) that provide stabilityand meaning to social behavior.

We define accountability as the capacity to ensurethat the agreed service objectives, performancestandards, procedures and payments contained in anirrigation service arrangement are complied with by thestakeholders involved.

Fig. 2 depicts such a concept of service provisionschematically. It shows the two important dimensionsthat have to be dealt with in any attempt to develop orimprove institutional arrangements – the externalinstitutional environment and the service arrange-

43

3. Institutional Requirements for Service Provision

ment. The following subchapters 3.1 and 3.2 describethese dimensions.

Fig. 2: Institutional Arrangements for Service Provision

3.1 The external institutional environment

We use the term “external institutional environment”to mean all the legal, policy, organizational, socio-economic and cultural factors which affect but are notpart of the direct service relationship itself.There maybe particular laws, prescriptions and regulations, thateffect the service provision.There may be conventions,political processes and other factors external to theirrigation system that have to be taken into account.How can we visualize such a multi-facetted construct?

In Figure 2, we represent the external institutionalenvironment by two bundles of external factors thatstrongly impinge on the service relationship.

44


1

1

2

3

2

3

Government

Service Provider

User/Civil Society

Service

Enabling Functions/ Subsidies

Governance Mechanisms betweenGovernment and Service Provider

Governance of Civil Society

Service Arrangement

� The first one relates to all the institutions thatgovern the relationship between the governmentand the service provider (whether this is anirrigation agency, a water user organization, aprivate firm or some other provider).

� The second one refers to all the institutionalmechanisms that govern the relationship betweenthe government and the service recipients (as partof civil society).

Based on such a perception, one may assume asupportive institutional environment, if there is (seeWorld Bank 1994 b):

In general,� A government with high legitimacy,� Accountability of political and official elements of

government (media freedom, transparent decision-making, accountability mechanisms),

� Respect for the rule of law,� A satisfactory public perception of the accountability

of civil servants,� A high degree of independence of the judiciary,� A satisfactory degree of administrative capacity in the

bureaucracy.

With respect to the relationship “government –service providers”, we consider that there is asupportive institutional environment if there is:

� A government which is competent to formulatepolicies and define its own role and corecompetencies with respect to service delivery,

� A well established framework for economic activity(laws on property rights, laws on companies(bankruptcy laws), banking, competition, foreigninvestment,establishment of regulatory bodies,etc.),

� Existence of formal mechanisms and informalchannels to facilitate communication between thepublic and the private sectors,

45

3.1 The external institutional environment

� Sufficient strength in the public procurementsystems (transparency of procedures, adoption ofbidding documentation, competitive bidding, stafftraining,etc.).

With respect to the relationship “government – users(civil society)”, we consider that there is a supportiveinstitutional environment where the followingconditions exist:

� Respect for human rights,� Political decision making based on strong

participation of relevant groups of civil society,� Microlevel accountability through beneficiary

participation in local decision making,� Easy access of users to fair legal procedures and other

conflict resolution processes,� A clear and transparent distribution of property

rights that is consistent with the intended servicedelivery system (see Module 5),

� A secure right of water user associations to organise.

3.2 Service arrangements

When trying to understand the exchange of servicesand returns within a network of interacting actors, theissue arises of how the system of exchanges needs tobe organised in order to be functional. A first steptowards answering this question is to focus on just twoexchange partners as shown in Fig.3.Both parties mustaddress the basic question of how to ensure thatservices and returns agreed upon are actually providedwithout one party taking undue advantage of the other.In other words,what needs to be established is a systemof agreements, contracts, rules and/or procedures that“govern” the exchange relationship. Such institutionsshould bring about sufficient coordination andmotivation to make the exchange happen to thesatisfaction of both parties.The same principle appliesin a network of interdependent actors where a

46


47

3.2 The service arrangement

multitude of bilateral exchange relationships need to becoordinated. In the following, we refer to the set ofcoordination mechanisms that organize a particularservice exchange as to a “service arrangement“.

The provision of any irrigation service involvesinteractions between service providers, servicerecipients, service payers, government policy makersand regulators. At the operational level, a serviceprovider may adjust gates,measure and distribute waterand apply sanctions against rule violators. Water usersmay provide payment for service and conveyinformation and recommendations or complaints to theservice provider.Indirectly,tax payers may subsidize thecost of irrigation in return for lower costs of food at themarket. Governments may provide subsidies forirrigation in return for compliance by water users withwater regulations. Service provision involves inter-dependent relationships among stakeholders.

For provision of commercial services, as depicted inFig. 3, the customer or client normally assumes threedifferent functions: he acts as the arranger, the payerand the consumer of the service – all at the same time.The commercial irrigation farmer who asks a privatefirm to install a pump for him,arranges this service,paysfor it and is the one who (hopefully) is able to makebeneficial use of this service. In this triple role, thefarmer has various possibilities to influence the service

Fig. 3: Commercial irrigation service provision: integration of

functions at the client side

Legend: see Fig. 4

48


Fig. 4: Irrigation service provision by government entity to water

users: splitting of functions at the client side

to be provided.As the arranger, the farmer selects theservice provider, assigns the terms of the service andauthorizes the provider to execute the necessary worksat his pumping station. Both in the selection processand in the formulation of the terms of contract thefarmer can make use of the authority to determine whatkind of service to buy. As the payer, he or she maywithhold or even refuse payment in case the service hasnot been provided according to the contractagreement.And as the consumer or user of the service,the farmer is the one who can express satisfaction ordissatisfaction during the provision process and ask formodifications. It is the direct interaction with the

service provider that allows the farmer to influence theprovision process according to his own needs. Ifexpectations cannot be fulfilled he/she can either alterthe terms of the contract or – at least for the next time– look for a different provider.

It is exactly this crucial feedback loop from theconsumer to the provider that is often missing in non-commercial service provision, as is shown in Fig. 4.Here, in many cases the functions of the arranger, thepayer and the user or consumer of the service are split.The service receiver (e.g.irrigation farmers that receiveirrigation water from a government agency on asubsidized basis) might neither be the full payer nor thearranger of the service that is being provided (see Fig.4). However, with well-established mechanisms ofcoordination and control between the different actorseven such arrangements can function effectively. Inpractice, however, this is often not the case. In thearrangement shown in Fig. 4 for example, we see, thatcoordination mechanisms between the water users andthe service providers,payers and arrangers are deficientor non-existent. It will hence be difficult to tailor theservice provision such that it takes the needs andpreferences of these water users into account. Eventhrough their payment decisions the water users canhardly influence the provider, there is no directconnection between the payment f0 of the water feesand the budget f1 and f2 provided for provision of theservice.The feedback loop that allows easy adjustmentof the service provision to the wants and needs of theclient is lost here. A service provision with deficientservice arrangements like these stands little chance offunctioning effectively.

Based on such considerations, we can say thatservice arrangements normally will be strong, if thereis:

� agreement among the involved parties upon clearobjectives of the service provision,

49

3.2 The service arrangement

� agreement upon well specified terms of the servicedelivery,

� agreement upon procedures and performancestandards(transparent,measurable and monitorable),

� a well established set of coordination mechanismsthat govern the relationships between the differentactors,

� a possibility for the client side to influence theprovision process, if so agreed,

� an accepted level of payments or returns and atransparent payment plan,

� ability and willingness of the client to pay,� a closed “feed-back loop”between service provision

and payment for that service,� the possibility for independent technical/financial

audits,� arrangement for transparent accounting procedures,� a mutually respected conflict resolution framework,� a high degree of client satisfaction with service

delivery.

50


Module 6 in Part Two of this Guide applies such considerations tomaintenance service provision. It presents guiding principles on how toanalyse and improve service arrangements for maintenance provision inirrigation.

4. Developing Institutionally ViableMaintenance Strategies

4.1 “Strategic Institutional Positioning”

The MAINTAIN concept emerged from the series ofcase studies and thematic papers, listed in the insidefront cover of this Guide and at the end of this volume.These studies showed that the inclusion of institutionalissues in the debate on maintenance has one particularconsequence: Since institutional contexts can vary sodrastically, it appears essential to differentiate basiccontexts that require different approaches tomaintenance improvements.

MAINTAIN responds to this requirement with“Strategic Institutional Positioning” (SIP). Such anapproach accommodates all the elements of theconceptual framework discussed above but permits thisto be done on the basis of a situational differentiation.

Since institutional environments can be quitediverse, approaches to solving maintenance problems– the “maintenance strategies” – will have to vary aswell. It goes without saying that maintenance efforts inthe Central Valley in California (see MAINTAIN Casestudy No.6) will have to have a different thrust and willconfront different institutional constraints than similarefforts in developing countries such as Jordan and India(see MAINTAIN Case studies No.3 and 5). This will beeven more true for a comparison with maintenanceactivities in least developed countries such as Haiti (seeGTZ-Publication Series 263). While it is impossible toprovide “recipies”for approaching every particular site-specific situation, we can define the characteristics ofsome basic contexts that will then give a certainorientation and guidance with respect to a particularcase in question.

SIP first attempts to visualize the important factors ofthe institutional environment in a two-dimensionalspace. This is an exercise which can only be

51

4. Developing Institutionally Viable Maintenance Strategies

What is the purpose

of Strategic

Institutional

Positioning (SIP)?

What is Strategic

Institutional

Positioning (SIP)?

52


implemented in a fairly crude way. SIP then requiresindicating the “position” of the particular irrigationsystem at hand in this two-dimensional space. Such aposition necessarily will be a rough approximation.However, the philosophy behind such a “quick anddirty” procedure is that it may be better to startmaintenance efforts with a rough idea of the rightstrategic option in a given institutional environmentthan to neglect the institutional context altogether,embark on ambitious programs to rectify theconsequences of deferred maintenance and then realizethat people just go on deferring maintenance activities.

To implement SIP,the evaluating team must considerthe context of the given irrigation scheme and assessthe two dimensions of “strength of servicearrangements” and “supportiveness of the externalinstitutional environment” discussed in sections 3.1and 3.2. Based on the resulting ratings a positioning ofthe particular problem situation can be discussed.Theimportant point here is not to attempt to deduct exactly

How can

“positioning” be

done?

Fig. 5: Strategic Institutional Positioning

the position of the given maintenance problem.RatherSIP should induce discussion and communicationbetween the involved stakeholders about theappropriate maintenance strategy to be followed. Inother words: the purpose of the instrument is not toprovide a mechanism that “automatically” leads to astrategy but to bring about discussions that createawareness about the relevant factors in the institutionalenvironment.

What are the basic maintenance strategies inextreme positions?

Such a positioning in one of the four quadrants of thestrategic space indicated in Fig. 5 will bring into thediscussion the following four basic strategy options:

This is the best case scenario. With both well-established service arrangements and a supportiveexternal institutional environment, the maintenancestrategy has to follow a very comprehensive approach,since the binding constraints that cause themaintenance problems may be of a very diverse nature.If the particular maintenance problem is positioned inan institutional environment that roughly correspondsto a position near the upper left corner of Fig. 5, thenthis might indicate that there are no or few institutionalcauses for the maintenance problems at hand. Theproblems, if any, are in most cases not related toinstitutional weaknesses and require a screening ofother problem domains, predominantly those of atechnical and an economic/financial nature. The factthat there are few or no institutional weaknesses relatedto the framework conditions or to the servicearrangement points to a very high level of performance,a level that will seldom be reached in developingcountries.A detailed analysis of the remaining problemswill be necessary and approaches of “AssetManagement”, as outlined in MAINTAIN ThematicPaper No.8,may be useful.

53

4.1 Strategic Institutional Positioning

Strategy option A:

Strong service

arrangements –

supportive

institutional

environment

Situations in the upper left corner of Fig. 5 as theymight occur in industrialized countries are not thefocus of this Guide. This is why detailed AssetManagement approaches are not a central feature of theset of Modules in Part Two of this book. However, forsituations at the fringes of the upper left quadrant,“Rapid Asset Appraisals”as introduced in Module 2 maybe a helpful tool.

Even if institutional problems are not the central issuein this quadrant,MAINTAIN Case Study No.4 indicateswith an example from the Central Valley of Californiathat institutional aspects can still acquire a high priorityin such situations. In the California case, with themandate of the US Bureau of Reclamation changingfrom a service providing to a more regulatory role,difficult adjustments of the service arrangementsbetween the water districts and their suppliers areneeded to cope with such dynamics.

However,positioning of the maintenance situation inthis quadrant indicates that a detailed assessment ofasset condition,importance and performance might beappropriate and that it may worth the effort to establishcomprehensive asset management programs as they areexplained in MAINTAIN Thematic Paper No.8.

This is why the main strategy for this quadrant canbe referred to as “Asset Management Strategies”.

This quadrant – the extreme opposite institutionalenvironment compared to case A – represents the worstcase scenario.Here neither the arrangements betweenthe provider and the users, nor the relationships withthe government or other essential supportive actors,arefounded on a sufficiently solid institutional base.Whatcan be the hope to establish a well-functioning andsustainable service delivery system for maintenanceunder such circumstances? Experience and commonsense indicate that such hopes will remain illusions.Theexample of the St.Raphael Irrigation System in Haiti inthe times of “Baby Doc” Duvalier (described in GTZ-publication No. 263) illustrates this point: all the

54


Strategy option D:

Weak service

arrangements –

unsupportive

institutional

environment

essential coordination mechanisms needed to establisha functioning water delivery and maintenance systemwere defunct. Endless cycles of deterioration andrehabilitation were the consequence.

What are the strategic options in such situations?

There are essentially three options:� One option might be to search for existing

coordination mechanisms in traditional smallcommunity systems. A good understanding of suchmechanisms might serve as a model for suchmechanisms in the context of the given irrigationscheme and then follow the ”enclave approach”described in strategy option C. In the Haiti case, athorough understanding of internal coordinationmechanisms employed in traditional small irrigationschemes in hill areas of Haiti might have helped toestablish functioning service arrangements in St.Raphael.

� If such models do not exist, then the major thrust ofany strategy needs to be sector wide efforts forinstitution building, time consuming and dependanton the political environment as such efforts may be.Simply developing maintenance manuals and trainingpeople to do maintenance – an approach oftenpursued in such situations – will be a predictablewaste of resources.

� A third option not often implemented so far is theoption of temporary external management takeoverby or a competent third party. In cases where thegovernment might consider the irrigation scheme inquestion to be vitally important for the nationaleconomy, it might opt for such a strategy, hoping totransfer the system back to local actors, onceinstitutional strengthening in line with the previousoption has been done. In situations of institutionalchaos and serious food shortages such an option maybecome a realistic point of discussion, in spite of allquestion marks that remain with respect to thefeasibility of the future transfer process.

55


Hence, positioning in this quadrant indicates thatnormally any maintenance approach needs to bepreceded by efforts at institution building .

This quadrant can be termed “Sector Strategies ofInstitution Building” .

This quadrant represents situations that are oftenfound in developing countries,where well functioningcommunity irrigation systems operate in externalinstitutional environments that are anything butsupportive.Formal water laws are either nonexistent ordo not consider local traditions sufficiently, the staff ofgovernment agencies are either badly trained orchronically underpaid or both,the regional and nationaljudiciary is heavily influenced by politicians, etc. Incontrast, local communities have retained strong socialcoherence and manage to operate their irrigationsystems,including maintenance provision,in a way that

56


Strategy option C:

Strong service

arrangements –

unsupportive

institutional

environment

Fig. 6: Situation specific maintenance strategies in different

institutional environments

fits well with their needs and preferences.The situationof small irrigation schemes in the Vilcanota Valley inPeru, described in MAINTAIN Thematic Paper No. 11,and of the Valle Alto Irrigation Scheme in Boliviamentioned in GTZ booklet No.263,correspond to suchcircumstances.

The strategy option here is clearly to prevent theunsupportive environment to encroach on andundermine the existing local institutions. This danger,however, is prevalent wherever external donors andnational or regional governments embark on theimprovement of such systems. Actors external to thelocal irrigation scheme now tend to define theobjectives to be pursued, push for modernization ofsuch systems that may not be in line with local rules andpractices, and support technical improvements. Thisoften occurs without the slightest understanding ofintricate local mechanisms of contractual governanceand how these are effected by the proposed changes.Strategic orientations therefore should follow a strictservice orientation,acknowledging the ownership andguidance of the local communities and being highlyaware that functioning local governance mechanismsare an asset that deserve protection and strengthening.

This quadrant is termed here the“Enclave approach”to water delivery and maintenance.

Predominant problems in such situations may relateto conflicts among the involved parties about theappropriate level of maintenance provision (seeModule 1) and problems of free-riding (see Module 8).In situations like these, maintenance efforts often donot take into account sufficiently the economicincentive situation of poor water users. Intensediscussions about the different institutional contextsmentioned in Module 3 may be of help here.

Positioning in this quadrant indicates situationswhere attempts are undertaken to reform servicearrangements for (water delivery and) maintenance ina supportive environment of government policies,laws

57


Strategy option B:

Weak service

arrangements –

supportive

institutional

environment

and regulations.Privatization of some sort or IrrigationManagement Transfer are the possibilities here. Thesituation of actual irrigation reforms in Turkey andMexico and to some extend in Andhra Pradesh in India,described in MAINTAIN Case Studies No.1, 5 and 6,correspond to the situation in this quadrant.The thrustof the strategy here,where reform is backed by genuinegovernment commitment, is directed toward thechange, build-up or strengthening of arrangements forservice delivery. The key challenge in this case is tobring about a genuine service orientation.This means,on the one hand,to transfer not only the role of the userand payer, but also much of the function of the“arranger”of water delivery and maintenance servicesto farmers. On the other hand, it requires a profoundshift in the self-perception of the agency staff from“patron”to “service provider.”Such changes imply wideranging structural reforms and can only be broughtabout when the incentives for irrigation staff are suchthat it “pays” for them to do so (see Module No. 10 ofthis Guide). Hence, this quadrant represents“Institutional Change Approaches”.

Most of the Modules of this Guide may be of help inthe situations falling into this quadrant.The core issuehere relates to the available options for institutionalarrangements for maintenance provision, irrigationservice provision and maintenance financing (seeModules 6 and 9). The preferred arrangements will bethose that allow for optimal coordination andmotivation of the involved actors. To approach sucharrangements, the identification of the involved actors(Module 4), an analysis of existing property rights andintended service functions (Module 5), and a soundunderstanding of available and functional mechanismsfor coordination of service relationships (Module 6) areessential steps.Existing service relationships need to beanalysed to identify potential problems of transparencyand accountability that open the door to highinefficiencies in service provision (Module 7). Andfinally, specific efforts need to be undertaken in order

58


to trace incentive deficiencies related to particularactors involved in the provision process (Modules and10).

4.2 Analysis and improvement of servicearrangements

One of the major insights of the MAINTAIN projectconcerns the nature of service provision – it can onlybe understood as an interactive processes with a varietyof contributors. Multiple actors have to invest money,time, physical and mental effort, attention and othersuitable resources into a production process thateventually generates the desired result:provision of themaintenance service.

The interaction between particular actors can beunderstood in terms of an exchange relationship: aparticular service may be provided in exchange for a feeor other tangible or intangible reward. For example:information may be delivered in exchange for a salary,or the service of representing a user organization at thepolitical level may be provided in exchange for honourand recognition of the representatives.

The hypothesis of this Guide is that non-existing ordeficient service arrangements are a prime cause formaintenance problems world-wide (not only inirrigation).Without functioning service arrangements,neither the coordination nor the motivation of theactors involved in maintenance provision will bepossible. However, discussions and analyses of servicearrangements in irrigation service provision in general,and in maintenance provision in particular, have notbeen part of appraisal, planning and evaluationprocedures in irrigation to date.

Based on the MAINTAIN concept presented aboveand on the practical experiences of the MAINTAINexcercises (see MAINTAIN Case Studies), werecommend the procedure summarized in Box 2 below.

59

4.2 Analysis and improvement of service arrangements

60


Part 2 of this guide presents the MAINTAIN Modulesand gives recommendations on how to adjust thesequence of steps and the use of the MAINTAINModules depending on the identified strategy.

Box 2: Sequence of steps to be followed when analysing andimproving service arrangements for maintenance provision:

1. Initiate discussions on “Strategic Institutional Positioning” (SIP) for theirrigation scheme in question involving the major stakeholders. Select astrategic orientation referring to the basic strategies A to D, describedabove.Depending on the chosen strategic orientation, make flexible use of thefollowing sequence of steps:

2. Identify the major actors involved in the provision of water delivery andmaintenance services (see Module 4).

3. Initiate discussions and reach agreement on level of key systemparameters to be achieved with maintenance efforts (see Module 1).Depending on the selected strategy, support the discussions onobjectives with an assessment of the infrastructural assets of theirrigation scheme and the related investments (see Module 2).

4. Initiate workshops with the involved stakeholders to answer thequestion: Who provides (or is supposed to provide) what kind ofservice or supporting service to whom? What returns orcompensations (payments) are made for the different services? Analysemajor deficiencies (see Module 4).

5. Initiate a participatory analysis of the existing property rights (seeModule 5).

6. Identify and document the existing authority system which defines theroles and functions of the involved stakeholders (see Module 5).

7. Initiate workshops and discuss the question: What are the external andinternal mechanisms that make sure that the provider delivers highquality service and that the client honors the service agreement andpays accordingly? Identify deficiencies and options for improvements(see Modules 6, 7 and 8).

8. In the same way as indicated for step 7, analyse the institutionalarrangements for Irrigation Financing (see Module 9).

9. In the same workshops discuss the question: What are the incentivesthat lead the provider to deliver high quality services? What are theincentives that induce the client to engage in the exchange relationshipand provide agreed-upon compensation? Identify deficiencies andoptions for improvements (see Module 10).

Part Two

Analysing and Improving Maintenance Provision

61

Table of Contents

Using the MAINTAIN-Modules to develop institutionally viable maintenance strategies and service arrangements 63

Module 1: Identifying service objectives and performance standards 65

Module 2: Rapid Asset Appraisal (RAA) 82

Module 3: Rapid Assessment of Economic Incentives for Maintenance 100

Module 4: The Service Interaction Analysis (SIA) 107

Module 5: Analysis of Property Rights and Authority Systems 119

Module 6: Analysing and Improving the Governance of Maintenance Provision 126

Module 7: Analysis of “Principal-Agent” problems 143

Module 8: Free Riding – A Central Governance Problem related to Maintenance 154

Module 9: Institutional Arrangements for Irrigation Financing 161

Module 10: Actor Specific ‘Incentive Profiles’ 172

Contents

62

Using the MAINTAIN-Modules todevelop institutionally viablemaintenance strategies and servicearrangements

When we widen the perception of maintenance toinclude institutional issues,we are obliged to face vastlydiverging institutional contexts. To respond to suchdiversity with attempts to look for the “one best way”to approach maintenance must be bound to fail.Hence,there is an urgent need to understand and discuss theresulting consequence that different situations requiredifferent maintenance strategies.The approach termed“Strategic Institutional Positioning”, introduced inChapter 4, can help to bring about such debates andawareness.

The package of independent Modules made availablein this Guide corresponds to such an approach.Depending on the prevailing strategy option andsubject to the dominant problem areas in a givensituation,different Modules may be referenced.

Table M0 summarizes the possible references tovarious Modules when discussing the different strategyoptions.

Subsequent to the title of each Module,we name themajor references and sources used as “SupportingDocuments”. Readers who are searching for detailedreferences and literature used should consult thesedocuments and the section “additional references” atthe end of this volume.

As stated before,the individual Modules may be usedindependently. However, this requires that the readeraccept some redundancies that are needed to minimizecross-references.

63

Using the MAINTAIN-Modules

64

Using the MAINTAIN-Modules

Table M0: Applying the MAINTAIN Guide in differing contexts

Type of Situation Most Probable Major Focus Major According to Fig. 5 Problem Areas of Analysis recommendedand 6, Chapter 4 Modules

A Few institutional � Asset Management Module 2 (RAA) isproblems; technical Approaches essential, selectedand economic/ Modules accordingfinancial to needstopics dominate

B Deficient institutional � Analysis of Main targetarrangements for coordination situation ofservice provision mechanisms this Guide.

� Accent on Principle-Agent problems Nearly all Modules

� Analysis of are applicable.incentive deficiencies

� Analysis of institutional arrangements forfinancing

C � Deficiencies in � Accent on Modul 1objective determination of (Service objectives)determination desired level

� Collective of service Modul 8goods problem � Solving of (Free-riding)

� Encroaching of free-riding external institutional problemsenvironment � Strengthening service General focuson local service orientation of official of Guidearrangements government agencies

D Deficient external General institutional This Guide with institutional strengthening focus on framework (problems maintenance with irrigation problems is not service are only sufficient here.secondary Change of focus problems) on general

support for policyand institutionbuilding required.

Module 1

Identifying service objectives and performance standards

(Supporting Documents: MAINTAIN Thematic PapersNo.5,8 and 12)

As we have discussed in section 2.2 of Part One ofthis Guide, a core aspect of a service orientation inmaintenance provision is clarity about the purpose ofmaintenance. After all, maintenance is not an end initself, but is a “secondary service” to the service ofirrigation water delivery. In the eyes of the farmers,maintenance efforts will only pay if they have asignificant effect on the farmer’s income level.However,establishing the connections between a specified levelof irrigation service and the maintenance required toensure that service is difficult. It is even more difficult,to establish the links between maintenance inputs andthe resulting increments in farm income.

Fig. M1-1. Maintenance causality chain

To shed light on these contingencies, it is necessaryto disaggregate the maintenance input/productionoutput equation. The schematic in Figure M1-1 depictsthe chain of connections between changes inmaintenance input and changes in net farm income.

Relationship 1 in the figure comprises the processof transforming maintenance funding into irrigation

65

Module 1

Why is it so

difficult to establish

objectives and

standards for

maintenance

provision?

∆ Maintenance ∆ AG Outputs

∆ Net Income∆ Irrigation Service

service. The effectiveness and efficiency with whichthese services are organized and provided defines thefunctional relationship between maintenance fundingand quality of irrigation service – the bang for the buckthat is provided.

Relationship 3 takes place within the farmingoperation. Here the farmer’s management skills, alongwith relative prices,determine the profit he makes.

Relationship 2 is a critical one for the purposes ofthis Guide, for it lies at the heart of what is “optimal” inoptimal maintenance. This relationship comprises the“contract”between the irrigation service provider andthe clients for the service.

Such a comprehension of the causality chain relatedto maintenance provision highlights the difficultiesencountered when trying to establish the involvedconnections.This explains why it is so seldom done.

There are two other reasons for the infrequentapplication of clear objectives and performancestandards for maintenance.First,budget constraints arenotorious in maintenance provision. Hence in manycircumstances, the available budget determines themaintenance program, rather than the maintenanceprogram determining the budget. Second, theidentification of the desired level of service provisionwill depend greatly on the institutional setup in whichthe irrigation system operates. Before we considerdifferent institutional contexts and the resultingdifferences in objective definition, let us define moreclearly the term “level of service”.

It is helpful to assess performance in the context ofa so-called “level of service provision”,both in terms ofthe water delivery service and in terms of the system’s

66

Module 1

What is the

“level of service”?

The purpose of this module is to introduce the concept of “level of service”and to give guidance with respect to objective determination based on such aconcept and considering different institutional contexts.

maintenance. The two are interlinked – maintenancecan affect water delivery and the operation of waterdelivery can affect maintenance.In general,a particularlevel of maintenance is a necessary prerequisite for thewater delivery service.

The level of service is a set of predefined operationalstandards that describe the quality of the water deliveryprovided to the water users (see MAINTAIN ThematicPapers 8 and 12).When defining a level of service, theobjective is to document and agree upon an officialstandard or norm against which the current serviceprovision can be compared. This also provides astandard when discussing the motivation of differentactors to contribute to the common objective (as willbe seen below).

The level of the maintenance service in irrigation isclosely linked to the level of service in water provision.Therefore the latter one needs to be well defined whendiscussing intended maintenance standards.

When we talk about a level of service, we need todifferentiate clearly between

� The official level of service – the level officiallystated and pursued by an irrigation organization or awater user association

� The potential level of service – the level which thetechnical system is able to provide;

� The desired level of service – the level of servicedesired by the involved stakeholders.

� The actual level of service – the level actuallyprovided.

In an ideal situation the official, the desired and theactual levels of service will coincide and they all willapproach the potential level fairly closely. However, inreality, it is essential to identify gaps between theselevels. To be able to do this we need to consider thedifferent types of levels separately.

67


68

Module 1

The “official” level of service

To identify the officially stated level of service bothfor water delivery and for maintenance,one may reviewofficial policy statements and other documents andinterview officials to determine the official view – bothwith respect to water delivery and with respect tomaintenance – about the following two questions:

� What services are supposed to be provided?� At what level or standard of service should these

services be provided,according to official policy andguidelines?Ideally,the level of water delivery services should be

expressed in terms of criteria that are relevant tofarmers such as (see MAINTAIN Thematic Paper No.8.In this paper, indicators are given for these criteria.)

� Area of command� Adequacy� Timeliness� Reliability� Security� Cost� Convenience� Flexibility

In an ideal case, the level of maintenance requiredwill be described in a so-called “asset and performancereport”that provides data on the function,performanceand condition of a particular piece of irrigation systeminfrastructure.

While such information might be available inirrigation schemes corresponding to situation Adescribed in chapter 4, which exist principally inindustrialized countries, many schemes in developingcountries have no explicit service objectives. Instead,managers – both of public and of farmer managedschemes – follow more input- or process orientedmethods of performance orientation. They stick to

certain administrative quotas or standard proceduresdepending on the resources available.

This scenario appears to be the common situation inpublic irrigation sectors of developing countries thatoften have to be positioned in the area betweenquadrants B and D of figures 5 and 6 in chapter 4. Buteven in these situations water service objectives arestated in general policy terms at the national or statelevel: such as targeted command area to be irrigated,discharges of water to be delivered for certain periodsat certain delivery points, or simply intentions toprovide water for two crops per year,one rice crop andone non-rice crop; or to provide five irrigations perseason for two seasons.These are further qualified andquantified at the scheme level. Sometimes serviceobjectives also include water delivery for domesticneeds, fisheries, and even rural industry.

The “potential” level of service

The potential level of service for water provisiondepends primarily on

� The type and design of the infrastructure� The capability of the involved actors to manage

operation and maintenance (O&M).

The potential level of service as determined by thetype and design of the infrastructure can be assessedthrough engineering studies. Such studies are not thesubject of this Guide.However, indicative relationshipssimilar to those given in table M 1-1 may be used forcanal systems. In this overview, different canal controland water delivery systems are graded according totheir level of service potential, the respective O&Mrequirements and costs.

The capability of the involved actors to manage(operation and) maintenance services refers to theircapability to take appropriate decisions with respect to

69


70

Module 1

Tab

le M

1-1:

Ind

icat

ive

rela

tio

nshi

p b

etw

een

leve

l of

inve

stm

ent,

can

al c

ont

rol,

leve

l of

serv

ice

and

O&

Mre

qui

rem

ents

and

co

sts

(Sou

rce:

Bur

ton

2000

in M

AIN

TAIN

The

mat

ic P

aper

No.

8)

Typ

eC

anal

Co

ntro

l sys

tem

1Fu

lly a

utom

ated

dow

nstr

eam

leve

l can

alco

ntro

l, fu

lly a

dju

stab

lean

d r

esp

onsi

ve t

ofa

rmer

dem

and

s

Wat

erD

eliv

ery

syst

em

Dem

and

Leve

l of

serv

ice

po

tent

ial

Very

hig

h, fu

lly r

esp

onsi

veto

farm

ers

dem

and

s fo

rw

ater

. H

ighl

y ef

ficie

nt in

wat

er u

se

O&

M r

equi

rem

ents

Low

sta

ffing

leve

ls d

ue t

oau

tom

atio

n, b

ut w

ork

forc

e ne

ed t

o b

e hi

ghly

skill

ed.

O&

M c

ost

s

Low

on

day

-to-

day

bas

is b

uthi

gh o

ver

time

as c

ontr

oleq

uip

men

t is

exp

ensi

ve.

Hig

hca

pita

l cos

t, lo

w O

&M

cos

t.

Cap

ital

inve

st-

men

t le

vel

Hig

h

Ind

icat

ive

O&

M c

ost

leve

l$/

ha

30

Po

ssib

lep

ote

ntia

lin

com

ele

vel

Hig

h

2M

anua

l con

trol

with

som

e au

tom

atio

n at

key

loca

tions

. D

isch

arge

mea

sure

men

t at

flow

div

isio

n an

d d

eliv

ery

poi

nts.

Arr

ange

d-

dem

and

H

igh,

res

pon

sive

to

farm

ers’

dem

and

s fo

rw

ater

tho

ugh

farm

ers

need

to

ord

er w

ater

inad

vanc

e. H

igh

inte

ract

ion

bet

wee

n se

rvic

e p

rovi

der

and

farm

er.

Hig

h st

affin

g le

vels

due

to

man

ual o

per

atio

n an

dne

ed fo

r m

easu

rem

ent

tom

atch

sup

ply

to

dem

and

.

Hig

h d

ue t

o co

st o

f O&

Mst

affin

g an

d a

ssoc

iate

dfa

cilit

ies

(offi

ces,

mot

orb

ikes

,et

c.).

Mai

nten

ance

cos

ts h

igh

to m

aint

ain

and

rep

lace

gat

esov

er t

ime.

Mod

erat

ely

high

60G

ood

3M

anua

l con

trol

thro

ugho

ut t

he s

yste

m.

Dis

char

ge m

easu

rem

ent

at fl

ow d

ivis

ion

and

del

iver

y p

oint

s.

Sup

ply

-d

eman

d

Mod

erat

e. S

upp

ly d

riven

with

irrig

atio

n se

rvic

ep

rovi

der

con

trol

ling/

allo

catin

g av

aila

ble

wat

erta

king

into

acc

ount

farm

ers

crop

pin

gp

atte

rns.

Rel

ativ

ely

low

inte

ract

ion

bet

wee

nse

rvic

e p

rovi

der

and

farm

er

Mod

erat

e st

affin

g le

vels

due

to

man

ual o

per

atio

nan

d n

eed

for

som

em

easu

rem

ent

to m

atch

sup

ply

to

dem

and

Mod

erat

e d

ue t

o O

&M

sta

ffing

and

nee

d fo

r so

me

O&

Mfa

cilit

ies.

Mai

nten

ance

cos

tshi

gh d

ue t

o ne

ed t

o m

aint

ain

cont

rol g

ates

.

Mod

erat

e40

Mod

erat

e

4M

anua

l con

trol

at

mai

nco

ntro

l poi

nts,

ung

ated

and

/or

pro

por

tiona

ld

istr

ibut

ion

at lo

wer

loca

tions

. Li

mite

dm

easu

rem

ent.

Sup

ply

Mod

erat

e, n

ot r

esp

onsi

veto

farm

ers’

dem

and

s,lim

ited

con

trol

ove

r w

ater

dis

trib

utio

n to

mat

chd

eman

ds.

Mod

erat

e to

low

sta

ffing

leve

ls d

ue t

o m

anua

lop

erat

ion,

tho

ugh

little

mea

sure

men

t

Mod

erat

e to

low

due

to

O&

Mst

affin

g an

d n

eed

for

som

efa

cilit

ies.

Mai

nten

ance

cos

tsm

oder

ate

due

to

need

to

mai

ntai

n m

ain

cont

rol g

ates

,ke

pt

low

er b

y lo

w-c

ost

cont

rol

at d

eliv

ery

poi

nts.

Low

30Lo

w

5Fi

xed

pro

por

tiona

lco

ntro

l sys

tem

, sup

ply

cont

rolle

d, n

otre

spon

sive

to

dem

and

.M

easu

rem

ent

at w

ater

sour

ce in

take

onl

y.

Sup

ply

Mod

erat

e to

low

, ot

resp

onsi

ve t

o fa

rmer

sd

eman

ds

for

wat

er b

utfa

rmer

s ca

n p

lan

ahea

dan

d a

dju

st c

rop

pin

gp

atte

rn t

o su

it su

pp

ly.

Inef

ficie

nt in

wat

er u

se.

Low

leve

l of s

taffi

ng, o

nly

low

ski

ll le

vels

req

uire

dLo

w d

ue t

o lo

w O

&M

sta

ffing

leve

ls a

nd t

o lo

w-c

ost

pro

por

tiona

l div

isio

n st

ruct

ures

.

Very

low

10S

ub-

sist

ence

planning,organizing,coordinating and controlling suchservices and with respect to incentive provision to allthose who participate. To assess such a capabilityrequires identifying relevant stakeholders,determiningthe services and support services they have to provideto each other,and to examining the viability of the rules,contracts, agreements and common practices thatgovern such relationships. Module No. 6 of this Guideprovides the basis for such assessments.

The “desired” level of service

The desired level of service corresponds to the actual“demand” of the stakeholders, especially the waterusers. However, the determination of this demandrequires different approaches in different institutionalcontexts.

Problems arise in cases where there are noinstitutional arrangements to make sure that allstakeholders, particularly the farmers, participate indetermining the desired level of provision. In suchcircumstances, the officially stated maintenanceobjective will not be identical with the level of servicereally desired.

The following differences in approach, referring tothe different scenarios explained in chapter 4 andFigures 5 and 6 (see pages 52 and 56) are important:

Situation A – the “Best case scenario”

We recall that in situations like these, the existingservice arrangements are well established andfunctioning and the external institutional environmentis supportive of such arrangements.

It is only in conditions of Situation A – as defined inchapter 4 – that the procedures to identify objectivesand standards for maintenance provision, as describedbelow, can be followed directly. In such a context,mechanisms are in place which ensure that the major

71


72

Module 1

stakeholders (or their representatives) can participatein the setting of objectives and standards. In the NesteSystem in Southern France (MAINTAIN Case Study No.2), this mechanisms is the “Comité Neste”. In thiscommittee,the major stakeholders decide upon servicelevels, tariffs, maintenance needs, and budgets. In theBroadview Water District in California (MAINTAIN CaseStudy No.5),the body to take decisions on maintenancelevels and budgets is the Board of Directors in whichthe farmers and the district management arerepresented and which is supported by a specialmaintenance committee (see Box M1-1).

In both cases the institutional conditions are suchthat stakeholder involvement in the objectivedetermination is fully garanteed. Under suchconditions, emphasis can be given to determining thedesired level of service and to discussing performancestandards as they are presented and discussed below.We refer to such approaches in the following as to the“standard approaches” to objective determination.

Box M1-1. Objective determination for maintenancein a California irrigation district

The Broadview Water District is a 4,000 hectare irrigation system inCalifornia’s Central Valley growing a variety of high value crops in a hot dryenvironment. It is governed by a board comprised of district landowners.Farmers value reliability of supply very highly in this environment, as a delayof several days in a scheduled irrigation can completely eliminate thegrower’s profit for the season.

When the current district manager took over several years ago, the systemwas seriously deteriorated. He convinced owners to impose on themselves aspecial maintenance fee to be used to bring the system back to the requiredhigh standards of reliability. The manager presents the board annually with aproposed budget for the coming year, and the board modifies and approvesit, automatically fixing the per hectare fees they will face for the year. (AfterCone, 2000 in MAINTAIN Thematic Paper No.11).

73


Situation C – the “Enclave Scenario”

In this situation, we have defined the maintenancestrategy of the “enclave approach”(see chapter 4).Thisrefers to farmer managed irrigation systems with astrong social coherence,where the system is functionaleven in an unsupportive external environment. Todetermine the desired level of service in suchcircumstances is seldom done in practice.Burton statesin MAINTAIN Thematic Paper No. 8, that “to theauthor’s knowledge little, if any, work has been doneon assessing farmers’ desired level of service insmallholder irrigation schemes in developingcountries”. Instead, the desired level of maintenanceservice is often defined from the outside with little orno intricate knowledge as to the internal institutionalarrangements that guide the decisions of the farmers.

As a consequence, this scenario requires a verycareful investigation into the really desired level ofservice.

Box M1-2. Objective determination for maintenance in Peruviansmallholder irrigation systems

In the mid 1990s, the German Government supported a program torehabilitate and improve more than a score of 100 to 200 hectare traditionalirrigation systems in the Peruvian Andes. When the program ended in 1997, amission was sent by the agency funding the work to evaluate the project. Themission found “serious organizational and technical deficits in the operationand maintenance of the irrigation systems as it was carried out by the waterusers associations.” This disturbing conclusion led to the dispatch of asecond mission early the following year. This second mission, including abroader set of perspectives, concluded that deficiencies existed, but that theywere not hydraulically significant. They noted farmers had demonstrated thecapacity to raise substantial resources to make emergency repairs, thatcritical problems were repaired immediately and that minor problems wereignored or delayed with good reason. This experience demonstrates theimportance of stekeholder objective determination for maintenance in smallscale schemes with strong social coherence as is the case in many regions ofthe Peruvian Andes. (After Urban, 2000 in MAINTAIN Thematic Paper No. 11.)

In doing this, some essentials of the standardapproach described below may be of use.The problemhere is that “demand for service”has to be understoodin the sense of an “effective demand”.The desired levelof service does not stand for an illusory level the farmersmay desire, but represents the level for which theyclients are prepared to pay or make the necessarycontri-butions.

In practice such an assessment may be difficult toimplement. Therefore, a “Rapid Economic Assessmentof Maintenance Needs” (REA) – as it is described inModule 3 of this Guide – may be implemented.We referto approaches of objective determination in suchcircumstances as to the “stakeholder objectivedetermination”.

Situation B – The “Hybrid Scenario”

In this case, the external institutional environmentappears quite supportive, but the existing servicearrangements are highly ineffective for irrigationservice provision.

In situations where service arrangements do notallow full and genuine stakeholder participation in theobjective determination,all attempts to do this withoutdeveloping or strengthening institutional arrangementswill lead to doubtful results. Maintenance may be amerit service here, i.e. a service provided by the stateto serve the collective well-being and supplied on thebasis of its own terms and conditions. However, theofficial level of service defined here has nothing to dowith a genuine desired level of service provision. Insuch cases the major focus of the objectivedetermination should be on the institutions andprocedures that lead to decision making aboutobjectives rather than on the details of the objectivedetermination itself.

74

Module 1

75


To the extent that this precondition is fulfilled,someof the essentials of the standard approach to objectivedetermination as it is presented below may be used.

We refer to this approach as the “merit-serviceapproach” to objective determination.

Situation D – The “Worst Case Scenario”

In this worst case scenario of institutional contexts,there are no actors who have a legitimate basis todecide on a level of service. Questionnaires andinterviews may identify a desired level of service butinstitutions and processes for stakeholder decisionmaking do not exist.

Here, it will not be possible to establish an agreedupon level of service unless farreaching reforms of theoverall institutional context have been brought about.

This is why we refer to this approach as to the“institution building approach” to objectivedetermination.

Fig. M1-2: Different approaches to objectice determination in

different institutional contexts.

A

C

B

D

Standard approach Merit service approach

Stakeholder objectivedetermination

Institution buildingapproach

Efforts to identify levels of service in such situations should be preceeded byor intimately linked with efforts to establish institutions and procedures thatallow such a participatory identification.

Determining the desired level of maintenance servicerequires establishing the following information:� Identifying different potentially feasible options for

the desired level of irrigation service, given theexisting infrastructure

� Identifying maintenance requirements related tothese levels of service,costs for the different levels ofservice, and the water users’ ability to pay

� Identifying the changes in income levels induced bythe different service levels comparing them with therespective costs.

As we have explained in the introduction to thisModule,such information is not easy to provide.In fact,even in industrialized countries, this information isseldom available.

There are techniques, employed in otherinfrastructure sectors, for establishing suchconnections explicitly, but these have rarely beenapplied in the irrigation sector.

These techniques generally fall under the heading of“Asset Management”.MAINTAIN Thematic Paper No.8describes these techniques and suggests ways thatcould be applied to irrigation system maintenance.Thispaper also presents a methodology for “Rapid AssetAppraisal” (RAA), an approach that is documented inModule 2 of this Guide.

However, even with the RAA methodology, theimpact of the individual asset performance on theoverall system performance and on the related incomedifferential is difficult to establish. Computer-basedsimulation models of system operation are sometimesuseful in specifying such relationships. However, suchapproaches may not be feasible in many circumstanceseven if they are positioned in basic situation A.

Depending on information needs,it may be sufficientin many cases to determine roughly the relationshipbetween the performance of particularly importantindividual pieces of infrastructure and the overallsystem performance. Importance relates primarily to

76

Module 1

Essentials of the

standard approach

to determining the

desired level of

service

77


Tab

le M

1-2:

Po

ssib

le p

erfo

rman

ce, c

ond

itio

n an

d im

po

rtan

ce r

elat

ions

hip

s (S

ourc

e: B

urto

n an

d H

all 1

999)

Per

form

ance

Co

ndit

ion

Imp

ort

ance

Pri

ori

tyE

xpla

nati

on

of

situ

atio

nC

ons

eque

nce

Goo

dG

ood

Hig

hLo

wN

o p

rob

lem

with

ass

et. P

erfo

rman

ce a

ndLo

w p

rob

abili

tyco

nditi

on a

re g

ood

, ind

icat

ing

that

ass

et

of s

truc

tura

l fai

lure

.is

new

and

in s

ervi

ceab

ility

gra

de

1 or

2.

Goo

dG

ood

Low

Low

No

pro

ble

m w

ith a

sset

. Per

form

ance

and

Lo

w p

rob

abili

tyco

nditi

on a

re g

ood

, ind

icat

ing

that

ass

et

of s

truc

tura

l fai

lure

. is

new

and

in s

ervi

ceab

ility

gra

de

1 or

2.

Goo

dP

oor

Hig

hH

igh

The

situ

atio

n is

haz

ard

ous

bec

ause

the

H

igh

pro

bab

ility

of s

udd

enas

set

is c

lose

to

failu

re, b

ut it

s go

od

stru

ctur

al fa

ilure

whi

chp

erfo

rman

ce m

ay p

rovi

de

a fa

lse

sens

e co

uld

hav

e hi

gh d

irect

and

of s

ecur

ity. H

igh

prio

rity

stat

us b

ecau

se

ind

irect

cos

t of

the

imp

orta

nce

ratin

g.co

nseq

uenc

es

Goo

dP

oor

Low

Low

The

situ

atio

n is

haz

ard

ous

bec

ause

the

H

igh

pro

bab

ility

of s

udd

enas

set

is c

lose

to

failu

re, b

ut it

s go

od

stru

ctur

al fa

ilure

whi

chp

erfo

rman

ce m

ay p

rovi

de

a fa

lse

sens

e co

uld

hav

e m

oder

ate

of s

ecur

ity. L

ow p

riorit

y st

atus

bec

ause

d

irect

or

ind

irect

cos

t of

the

imp

orta

nce

ratin

g.co

nseq

uenc

es.

Poo

rG

ood

Hig

hH

igh

Hig

h p

riorit

y st

atus

as

per

form

ance

is

Eng

inee

ring

asse

ssm

ent

low

and

imp

orta

nce

is h

igh.

Con

diti

on is

re

qui

red

to

iden

tify

the

good

ind

icat

ing

that

per

form

ance

is

pro

ble

m c

ausi

ng t

he p

oor

affe

cted

by

som

ethi

ng o

ther

tha

n co

nditi

on.

per

form

ance

.

Poo

rG

ood

Low

Low

Low

prio

rity

stat

us s

ince

imp

orta

nce

is lo

w.

Eng

inee

ring

asse

ssm

ent

req

uire

dC

ond

ition

is g

ood

ind

icat

ing

that

to id

entif

y th

e p

rob

lem

per

form

ance

is a

ffect

ed b

y so

met

hing

caus

ing

the

poo

rot

her

than

con

diti

on.

per

form

ance

.

Poo

rP

oor

Hig

hH

igh

Hig

h p

riorit

y st

atus

as

per

form

ance

and

H

igh

pro

bab

ility

of s

udd

enco

nditi

on a

re p

oor

and

imp

orta

nce

is h

igh.

st

ruct

ural

failu

re w

hich

cou

ld

This

ind

icat

es t

hat

the

asse

t ha

s fa

iled

and

ha

ve h

igh

dire

ct a

nd in

dire

ctis

in s

ervi

ceab

ility

gra

de

4 or

5.

cost

con

seq

uenc

es.

Poo

rP

oor

Low

Low

Low

prio

rity

stat

us a

s im

por

tanc

e is

low

. H

igh

pro

bab

ility

of s

udd

enH

owev

er, t

he p

oor

per

form

ance

and

con

diti

on

stru

ctur

al fa

ilure

whi

ch c

ould

in

dic

ate

that

the

ass

et h

as fa

iled

, or

is a

bou

t to

ha

ve m

oder

ate

dire

ct o

r in

dire

ct

fail,

and

is in

ser

vice

abili

ty g

rad

e 4

or 5

.co

st c

onse

que

nces

.

78

Module 1

the asset’s function, position in the irrigation ordrainage network, and its replacement value. A riverdiversion weir is more important than the secondarycanal head regulator,for example,because of its centralfunction in diverting and controlling inflow to thescheme, its position at the head of the system, and its(usually) significant replacement cost.

Through engineering studies, the cost database formaintaining or enhancing the condition/performanceof each type of asset (river weir, canal head regulator,aqueduct,culvert, etc.) can be ascertained and appliedto the asset condition/performance of each asset. Inthis way the cost of maintaining or enhancing thecondition/performance of the irrigation and drainagesystem is determined. An indication of the possiblerelationship between the condition, performance andimportance is presented in Table M1-2.

Under ideal conditions,especially in situations of thetype A, studies may also identify the anticipatedimprovements in performance benefits arising fromdifferent levels of investment. However, in most cases,this will be neither practical nor feasible.It may then behelpful to refer to the tool of “Rapid EconomicAppraisal”presented in Module 3.

The following criteria for the water delivery serviceare often particularly stressed by farmers (after Burtonand Hall 1999):

The criteria in the table have been ranked toemphasise the fact that farmers have different levels ofpriority for the various criteria,and will be prepared toforego some and not others. The ranking is schemespecific and may vary between farmers.Obtaining theserankings is not easy, but is essential if the desired levelof service is to be defined.

In agency-farmer managed systems,a similar table ofcriteria and priorities can be constructed for theirrigation service provider. Here, the priority of thecriteria may be quite different.The service provider mayhave to compromise output as a consequence of

Farmers’ criteria

79


limitations related to inputs (river flow pattern) andprocesses (control infrastructure).

In this case,the performance assessment will have tobe done at three levels additional to the farmers’ levelmentioned before: scheme, system and statutory.

Scheme level relates to overall performance of thescheme and uses criteria and performance indicatorsthat produce an overall assessment for the scheme.

System level relates to the irrigation network anduses criteria and performance indicators that relatemainly to the inputs and the processes of waterconveyance.Adequacy and timeliness and command inthis respect relate specifically to input (at water source)and process (throughout the network), not to output.Equity, efficiency (conveyance and pumping whereused) and financial cost are key criteria at this level.

Statutory requirements are those, such as drainageoutfall from irrigation schemes into rivers,which mightbe stipulated by law.

Table M1-4 shows a “Servicability Matrix” that mayhelp guide the establishment of relevant criteria and forthe preparation of negotiations on the desired level ofservice. A final, but extremely important criterion for

Table M1-3: Possible criteria for assessing level of serviceprovision (of irrigation water supply) from farmers’ perspective(Source: Burton and Hall, 1999)

High Priority Moderate Priority Low Priority

� Command (water level) � Cost � Efficient

� Adequacy � Quality � Equitable (Fair)

� Timeliness � Convenience � Safety

� Reliability � Flexibility

� Security

80

Module 1

Tab

le M

1-4:

Po

ssib

le Ir

rig

atio

n S

ervi

ceab

ility

Mat

rix:

cri

teri

a an

d c

lass

ifica

tio

ns (

afte

r B

urto

n an

d H

all,

1999

)

SC

HE

ME

PE

RFO

RM

AN

CE

SY

ST

EM

PE

RFO

RM

AN

CE

Gra

de

Pro

duc

tivi

tyC

rop

pin

g

Co

stE

ffic

ienc

yA

deq

uacy

and

Co

mm

and

Eq

uity

E

ffic

ienc

yC

ost

Inte

nsit

y(E

cono

mic

)(R

eso

urce

T

imel

ines

so

f su

pp

ly(F

inan

cial

)us

e)

1G

reat

er t

han

Gre

ater

tha

n Ve

ry h

ighl

yVe

ry h

igh

Ad

equa

te a

ndTa

rget

leve

lsW

ater

E

ffici

ency

leve

lsH

ighl

y vi

able

90%

of p

oten

tial

90%

of p

oten

tial

econ

omic

timel

y at

all

times

mai

ntai

ned

at

dis

trib

utio

nm

atch

tar

get

all t

imes

equi

tab

leva

lues

270

-89%

of

70-8

9%

of

Hig

hly

Hig

hG

ener

ally

Targ

et le

vels

Dis

trib

utio

nE

ffici

ency

leve

lsVe

ry v

iab

lep

oten

tial

pot

entia

lec

onom

icad

equa

te a

nd t

imel

yge

nera

lly

gene

rally

gene

rally

mai

ntai

ned

eq

uita

ble

adeq

uate

350

-69

% o

f50

-69

% o

fM

oder

atel

yM

oder

ate

Ad

equa

te a

ndTa

rget

leve

lsD

istr

ibut

ion

Effi

cien

cy le

vels

Mod

erat

ely

pot

entia

lp

oten

tial

econ

omic

timel

y on

ave

rage

mai

ntai

ned

eq

uita

ble

adeq

uate

viab

leon

ave

rage

on a

vera

geon

ave

rage

430

-49

% o

f 30

-49

% o

fM

argi

nally

Low

Freq

uent

lyTa

rget

leve

lsD

istr

ibut

ion

Effi

cien

cy le

vels

Low

via

bili

typ

oten

tial

pot

entia

lec

onom

icin

adeq

uate

/unt

imel

yfr

eque

ntly

fr

eque

ntly

freq

uent

lyno

t m

aint

aine

d

ineq

uita

ble

inad

equa

te

5Le

ss t

han

29%

Le

ss t

han

29%

Une

cono

mic

Very

low

Com

ple

tely

Leve

ls n

otW

ater

E

ffici

ency

leve

lsN

ot v

iab

leof

pot

entia

lof

pot

entia

lin

adeq

uate

/ un

timel

ym

aint

aine

d.

dis

trib

utio

nun

acce

pta

ble

is in

equi

tab

le

FAR

ME

RS

’ RE

QU

IRE

ME

NT

SS

TAT

UT

OR

Y R

EQ

UIR

EM

EN

TS

Gra

de

Ad

equa

cy a

nd

Co

mm

and

Rel

iab

ility

Sec

urit

yW

ater

Wat

erlo

gg

ing

Hea

lth

Env

iro

nmen

tT

imel

ines

so

f su

pp

lyo

f sy

stem

qua

lity

and

Flo

od

ing

and

Saf

ety

Gra

de

1A

deq

uate

and

Ta

rget

leve

lsFu

lly r

elia

ble

No

risk

No

Non

eC

omp

lies

Com

plie

stim

ely

at a

ll tim

esm

aint

aine

d a

t al

l tim

esof

failu

reco

nstr

aint

s

Gra

de

2G

ener

ally

ad

equa

te

Targ

et le

vels

Gen

eral

lyS

ome

risk

Som

eS

ome

Not

Mild

haz

ard

and

tim

ely

gene

rally

mai

ntai

ned

re

liab

leof

failu

reco

nstr

aint

inci

den

ceap

plic

able

Gra

de

3A

deq

uate

and

tTa

rget

leve

lsR

elia

ble

on

Mod

erat

e ris

kM

oder

ate

Mod

erat

eN

otM

oder

ate

imel

y on

ave

rage

mai

ntai

ned

on

aver

age

aver

age

of fa

ilure

cons

trai

ntin

cid

ence

leve

lsap

plic

able

haza

rd

Gra

de

4Fr

eque

ntly

inad

equ.

Ta

rget

leve

ls fr

eque

ntly

Freq

uent

lyH

igh

risk

Ser

ious

Ser

ious

Not

Ser

ious

and

/or

untim

ely

not

mai

ntai

ned

un

relia

ble

of

failu

reco

nstr

aint

inci

den

ce le

vels

app

licab

leha

zard

Gra

de

5C

omp

lete

ly in

adeq

uate

Com

man

d le

vels

Com

ple

tely

Faile

d o

rQ

ualit

y fa

tal

Una

ccep

tab

le

Non

-N

on-

and

/or

untim

ely

not

mai

ntai

ned

.un

relia

ble

failu

re im

min

ent

to a

gric

ultu

ral

inci

den

ce le

vels

Com

plia

ntC

omp

liant

pro

duc

tion

81


the desired level of service is the ‘willingness to pay’.This willingness depends on available incentives butabove all on the economic assessement thatstakeholders do with respect to different levels ofservice.However,as we have stated repeatedly, the linkbetween a certain level of maintenance, the resultingeffect on the water delivery service, and the ensuingincremental changes in production and income levelsis extremely difficult to establish. Nevertheless,stakeholders develop perceptions of costs and benefitsrelated to different levels of maintenance serviceprovision.Module 3 provides some guidance on how toevaluate such assessements.

Assuming that the official and the desired levels ofservice for water delivery do not surpass the potentiallevel of service the system can provide, then gapsbetween the level of service actually provided and theofficial and desired levels of service can be quite indi-cative of the problems encountered.Such a “goal incon-gruence”can be interpreted as shown in Table M1-5.

Gaps between levels

of service – what do

they tell us?

Table M1-5: Gaps between levels of service Degree of goal congruence Indication Strategic orientation Occurence in

of further actions particular situations

Official = desired = actual l.o.s. No O&M Best practice scenario: Most probable in situationsproblems maintain actual l.o.s. A and C described

in chapter 4

Official ≠ desired = actual l.o.s. System has Adjust official l.o.s. Frequent situation inadjusted to Prevent external community systemsdesired l.o.s. official views to with external support

impinge on system (Situation C in chapter 4)

Official = desired ≠ actual l.o.s. No problems of goal Maintain official l.o.s. Frequent situation inincongruence but Analyse system to community systemsother problems. High identify causes for with internal problemslevel of stakeholder suboptimal l.o.s. (Situation C in chapter 4)coordination.

Official ≠ desired ≠ actual l.o.s. Problems of goal Review of desired l.o.s. Most frequent situationincongruence plus and careful analysis in developing countryother problems of the service system irrigation (Situations B, C

and D in chapter 4)

Official ≠ desired l.o.s. Centrally administered Discuss pros and cons Special case of ‘enclave’system with strong of system change to situation (e.g. state farm)

Official = actual l.o.s. management desired l.o.s. (Situation C in chapter 4)

l.o.s. = level of service

Module 2

Rapid Asset Appraisal (RAA)

(Supporting Documents: MAINTAIN Thematic PapersNo.8 and 11)

The core idea behind “asset management” is therecognition that infrastructure is supposed to providea certain service from which benefits can be derived.Canals allow the delivery of irrigation water,drains canbe used to evacuate drainage water,roads serve to allowor facilitate transport and mobility. Maintaining orenhancing that service providing function results insustained or enhanced benefits, either financial orsocial.

This idea has farreaching consequences: in fact, itimplies that the quality of maintenance is closely relatedto the level of service that one intends to maintain.Thisis an important shift in maintenance philosophy,namelya shift from an input to an output orientation.Maintenance – not only in irrigation – traditionally hasbeen an input oriented excercise. Maintenance staffand/or contractors used to face (and still face) amultitude of specifications on materials and proceduresbut are not required to guarantee the infrastructure’sability to provide the intented level of service.Maintenance manuals present prescriptions on how,when and with what means to do maintenance.However, they normally do not help to linkmaintenance provision to a given output level,e.g. to apredetermined “level of water delivery service” inirrigation.

82

Module 2

What is “Asset

Management”?

Following this line of thinking, Asset Management (AM) is a structured andauditable process for planning maintenance of and investment ininfrastructure to provide users with a sustainable and defined level of service.

Asset Management establishes connections betweenmaintenance and maintenance expenditure on onehand, and asset condition and system performance onthe other. It does this by establishing the followingchain of connections (Box M2-1).

This is obviously not a simple process and is very dataintensive. Most applications are computer-based andgeneric commercial software for AM is available.Computer-based simulation models of system operationare also useful in specifying some of the relationshipsinvolved, that between individual asset performanceand system performance in particular. The nearestwidespread application of this approach to theirrigation sector is in public water supply systems,where it is used extensively in the UK and elsewhere.

83


Given these preconditions the applicability of Asset Management approacheswill be limited in most cases to institutional contexts such as those describedwith “Situation A” (“Best Case Scenario/Asset Management Strategies”) inchapter 4. However, in other situations a simplified form of AssetManagement, the “Rapid Asset Appraisal”, presented below, may be appliedas a supporting tool.

Box M2-1. Essential steps in Asset Management

Determing the desired level of service

Determining the current extent, condition, value and performance of individual assets

Relating individual asset extent and performance to system performance

Determining the cost of maintaining or enhancing performance of each type of system asset

In looking at maintenance of irrigation and drainagesystems it is important to be aware that the conditionand performance of the infrastructure is a function ofits:

� Design� Construction� Operation� Maintenance

The quality of the design and construction influencethe rate at which the infrastructure deteriorates andhow it performs its intended function. How theinfrastructure is used and operated can affect itscondition and performance, as can the level ofmaintenance.

In this Module,we do not deal with matters of design,construction and operation. In practical cases onewould have to investigate whether or not and to whatextent these factors influence condition andperformance of the infrastructure. Here, we focusexclusively on maintenance. Design and constructiondo enter the framework, however, by setting a limit onthe “potential level of service”discussed in module 1.

In relation to maintenance, the condition andperformance of the infrastructure is influenced by thelevel of:

� day-to-day maintenance � annual maintenance � emergency maintenance� deferred maintenance � capital replacement.

84

Module 2

What information is

needed for a Rapid

Asset Appraisal?

The purpose of this Module is to introduce a procedure for Rapid AssetAppraisal. Such a procedure can help to establish basic connections betweenmaintenance and system performance, a link that is crucial in maintenanceservice provision.

Information on the first three components isrelatively straightforward to obtain from records keptby the irrigation service provider and discussions withservice provider staff.The information required relatesto the expenditure,type and extent of the work carriedout.

Information on deferred maintenance (which is anaccumulation of failure to adequately carry out allrequirements under the first three categories) isdifficult to obtain, as records of total (outstanding)maintenance requirements are often not kept.Failure tocarry out necessary maintenance work “mines“ theasset base, resulting in system deterioration.

The final category, capital replacement, representsexpenditure to replace assets as they reach the end oftheir useful life, or become obsolete. Failure toadequately maintain the assets during their lifetime canobviously lead to a more rapid deterioration and areduced life expectancy. This category is often notconsidered in maintenance studies.

For the fourth and fifth maintenance categoriesidentified above a detailed study is required of the assetbase to assess it current condition and level ofperformance. During an appraisal of the institutionalissues related to maintenance there is not the time tocarry out a detailed study of the asset base andsimplified measures have to be used.

For RAA, data need to be collected to determine thecurrent condition and level of performance of theinfrastructure, including the following:

� the extent of the asset base� the condition of the assets� the performance of the assets� the importance of individual assets

Where the asset base is maintained in a good state ofrepair, and assets are replaced as they reach the end oftheir useful life, as is the case with the Neste System inSouthern France (MAINTAIN Case Study No.2), such a

85


What are the essential

steps to be taken

when implementing

a RAA?

study might not be required, except to establish thatthese conditions do prevail. In this case, theexpenditures made for maintenance reflect the truecost of maintaining the system over time.

The steps required to carry out a RAA are summarisedin Table M2-3 (see pages 92/93) and discussed below.

� System overviewDuring this step the infrastructure is inspected anddiscussions held with farmers and service providerstaff on their perception of scheme performance andlevels of service provision. Note should be taken ofthe appearance of the crops, crop yields, marketing,soil conditions, farming practices, etc.

� Obtain general background dataThe next step is to obtain background information onthe scheme. This will include maps and aerialphotographs (if available), and records of croppingover recent years. This data will form the backboneof subsequent data collection and analysis.The extent and quality of the data available willprovide insight into the standard of management,operation and maintenance on the scheme.

� Obtain and process detailed system performancedataDetailed data are collected to enable an assessmentto be made of the performance of the irrigationscheme and to identify potential areas of concern.Secondary data are required for this analysis, sincethere is usually not time to collect primary data.

� Identify current and potential performance levelsand current and desired levels of service provisionAnalysis of the data will help to identify the level ofperformance of the scheme. Application of Module1 of this Guide can be used to help determine thedesired level of service. From the analysis of theactual performance and the system configuration,anassessment can be made of the potential

86

Module 2

performance level, leading to identification of thegaps and current constraints to production. Thisanalysis is a major activity.The degree to which it iscarried out depends upon the context, the timeavailable and the experience of those performing theassessment. Arising from the analysis will be a firstrough assessment of current and potentialperformance of the scheme.

� Obtain maintenance dataInformation needs to be collected on the extent andtype of maintenance work carried out and theexpenditure on maintenance. This information isrequired for a period of at least 5 years, if possible, toassess trends. From analysis of the data it will bepossible to form an opinion on the maintenancesituation and its likely impact on the condition andperformance of the infrastructure.

� Determine extent of existing asset database, stratifyand select sample baseThe ease with which an asset database can beobtained varies from scheme to scheme. Manyschemes have asset inventories and schematicdiagrams that provide information on the asset baseand the location of assets. Some schemes have as-built construction drawings which can proveinvaluable. Having such records obviously simplifiesthe task of preparing an asset database. The validityof the database can be assessed during the assetsurvey. In large schemes, a stratification of the assetscan be carried out and the number of samples setsand their size determined,making use of the databaseand field inspections. (For further details see Annex1 of MAINTAIN Thematic Paper No.8).

� Carry out asset surveyTo save time it will be necessary to map out thelocation of the assets selected for inspection and tomove through the system inspecting them.In almostall cases the headworks will be a one-off assessment,

87


88

Module 2

Table M2-1: Performance services of physical components ofirrigation and drainage schemes (Source : Burton 2000 in MAINTAIN Thematic Paper No. 8)

Component Levels Performance service

Canals Primary To convey waterSecondaryTertiaryQuaternary

Drains Primary To remove water from the fieldSecondaryOn-farm

River weir Main canal To divert and control irrigation supplies

Headworks Main canal To take in water to the main canal. This may be a groupintake of structures, including a river weir, head regulator,

settling basin, and measuring structure, or one structuresuch as a pump station.

Pump station Main canal To lift water to command level for irrigation. To removeMain drain water from drainage channels which are below river level

Settling basin Main intake canal To settle out sediment

Cross regulator Primary and To raise and maintain water surface at design elevationsecondary canals

Head regulator Primary, secondary To regulate discharge entering a canaland tertiary canals

Measuring Primary, secondary To measure discharge for operational purposesstructure and tertiary canals

Aqueduct All levels of canal To pass canal over an obstruction (another canal, a drainage channel, etc)

Culvert All levels of canal To pass canal or drain under an obstruction (road,or drain drainage channel, etc)

Drop structure All levels of canal To „drop” the canal or drain bed level in a safe manner. or drain Used to slacken canal or drain slopes on steep land

Escape All levels of canals To release water from a canal into the drainage networkstructure in the event of oversupply or under-utilisation.

Syphon All levels of canals To pass the canal below an obstruction such as a road underpass or drainage channel.

Distribution box Quaternary canal To distribute water between quaternary channels

Night storage Main canal or To store irrigation water during the night for releasereservoir on-farm during the day. Main canals can thus operate 24

hours/day whilst lower order canals can be operatedduring the daytime.

Tubewell On-farm To abstract groundwater for irrigation. Often used inconjunction with surface water system

Bridges Road bridges To allow human and animal traffic over the canal or drainFoot bridges

Roads Inspection roads To gain access to the irrigation system and villages.Access roads For inspection and maintenance

the asset survey can start there and proceeddownstream. During the asset survey it is valuable tobe aware of the system as a whole and note anyfeatures/factors which might influence schemeperformance.The asset survey involves the following steps.

– Defining the function – i.e. the infrastructureservice – performed by a certain asset.As an orientation,a list of performance services ofdifferent components of irrigation and drainageschemes is given in table M2-1 on the previouspage (repeating table 3 in Part One of this Guide)

– Condition grading of the asset (see Annex to Module 2)

– Performance grading of the asset(see Annex to Module 2)

– Importance grading of the asset (see Annex to Module 2)

Importance relates primarily to the asset’s function,position in the irrigation and drainage network, andits replacement value.A river diversion weir is moreimportant than a secondary canal head regulator, forexample, because of its central function in divertingand controlling inflow to the scheme, its position atthe head of the system and its (usually) significantreplacement cost.With such information, standard proforma can bedrawn up, or notes made in a notebook using a datacollection checklist for each asset type.An exampleof a data collection proforma is given in Table M2-4(see page 94). Examples of how to go about withasset condition/performance/importance gradingare given in the Appendix to this Module.

� Formulate asset condition and performance reportOnce the sample sets of assets have been surveyed,the data set can be extrapolated to characterize thewhole population and a picture obtained of thecondition and performance of the scheme’s

89


90

Module 2

Tab

le M

2-2

(iden

tica

l wit

h ta

ble

M1-

2): P

oss

ible

per

form

ance

, co

ndit

ion

and

imp

ort

ance

rel

atio

nshi

ps

(Sou

rce:

Bur

ton

and

Hal

l 199

9)

Per

form

ance

Co

ndit

ion

Imp

ort

ance

Pri

ori

tyE

xpla

nati

on

of

situ

atio

nC

ons

eque

nce

Goo

dG

ood

Hig

hLo

wN

o p

rob

lem

with

ass

et. P

erfo

rman

ce a

nd

Low

pro

bab

ility

cond

ition

are

goo

d, i

ndic

atin

g th

at a

sset

is

of s

truc

tura

l fai

lure

. ne

w a

nd in

ser

vice

abili

ty g

rad

e 1

or 2

.

Goo

dG

ood

Low

Low

No

pro

ble

m w

ith a

sset

. Per

form

ance

and

Lo

w p

rob

abili

tyco

nditi

on a

re g

ood

, ind

icat

ing

that

ass

et is

of

str

uctu

ral f

ailu

re.

new

and

in s

ervi

ceab

ility

gra

de

1 or

2.

Goo

dP

oor

Hig

hH

igh

The

situ

atio

n is

haz

ard

ous

bec

ause

the

ass

etH

igh

pro

bab

ility

of

is c

lose

to

failu

re, b

ut it

s go

od p

erfo

rman

ce

sud

den

str

uctu

ral

may

pro

vid

e a

fals

e se

nse

of s

ecur

ity. H

igh

failu

re w

hich

cou

ldp

riorit

y st

atus

bec

ause

of t

he im

por

tanc

e ra

ting.

have

hig

h d

irect

and

ind

irect

cos

t co

nseq

uenc

es

Goo

dP

oor

Low

Low

The

situ

atio

n is

haz

ard

ous

bec

ause

the

ass

et

Hig

h p

rob

abili

ty o

f is

clo

se t

o fa

ilure

, but

its

good

per

form

ance

su

dd

en s

truc

tura

lm

ay p

rovi

de

a fa

lse

sens

e of

sec

urity

. Low

fa

ilure

whi

ch c

ould

p

riorit

y st

atus

bec

ause

of t

he im

por

tanc

e ra

ting.

have

mod

erat

e d

irect

or

ind

irect

cos

t co

nseq

uenc

es.

Poo

rG

ood

Hig

hH

igh

Hig

h p

riorit

y st

atus

as

per

form

ance

is lo

w

Eng

inee

ring

asse

ssm

ent

and

imp

orta

nce

is h

igh.

Con

diti

on is

goo

d

req

uire

d t

o id

entif

y th

e in

dic

atin

g th

at p

erfo

rman

ce is

affe

cted

by

pro

ble

m c

ausi

ng t

he p

oor

som

ethi

ng o

ther

tha

n co

nditi

on.

per

form

ance

.

Poo

rG

ood

Low

Low

Low

prio

rity

stat

us s

ince

imp

orta

nce

is lo

w.

Eng

inee

ring

asse

ssm

ent

Con

diti

on is

goo

d in

dic

atin

g th

at p

erfo

rman

cere

qui

red

to

iden

tify

the

is a

ffect

ed b

y so

met

hing

oth

er t

han

cond

ition

.p

rob

lem

cau

sing

the

poo

r p

erfo

rman

ce.

Poo

rP

oor

Hig

hH

igh

Hig

h p

riorit

y st

atus

as

per

form

ance

and

H

igh

pro

bab

ility

of s

udd

en

cond

ition

are

poo

r an

d im

por

tanc

e is

hig

h.

stru

ctur

al fa

ilure

whi

ch

This

ind

icat

es t

hat

the

asse

t ha

s fa

iled

and

is

coul

d h

ave

high

dire

ct a

nd

in s

ervi

ceab

ility

gra

de

4 or

5.

ind

irect

cos

t co

nseq

uenc

es.

Poo

rP

oor

Low

Low

Low

prio

rity

stat

us a

s im

por

tanc

e is

low

. How

ever

, H

igh

pro

bab

ility

of s

udd

enth

e p

oor

per

form

ance

and

con

diti

on in

dic

ate

that

the

st

ruct

ural

failu

re w

hich

cou

ld

asse

t ha

s fa

iled

, or

is a

bou

t to

fail,

and

is in

ha

ve m

oder

ate

dire

ct o

r in

dire

ct

serv

icea

bili

ty g

rad

e 4

or 5

.co

st c

onse

que

nces

.

infrastructure. In a comprehensive Asset Manage-ment appraisal, the assessment of the impact of thecurrent condition and performance of individualassets on the overall performance of the scheme hasto be made once all the data has been collected. Thisassessment is not easy, and some subjectivejudgement will be required to make the assessment.Recent studies by El-Askari (El-Askari 1999; GICC1998) have shown the significant value of usinghydraulic modelling to aid such assessment. In thesestudies El-Askari used hydraulic modelling to identifylinkages between asset performance in one part ofthe irrigation system with impacts at other locations.Amongst others, the impact on downstream waterdelivery of sediment levels within canal sections wasinvestigated,as was the impact on downstream waterdelivery of damaged or poorly maintained controlstructures.

In cases,where such data collection and modellingwork cannot be done,it may be sufficient to establishpossible performance-condition-importance relation-ships. Such relationships will allow establishingpriorities for maintenance intervention, even if theexact impact of asset conditions on overallperformance of the scheme is not known.

Repeating table M1-2,an indication of the possiblerelationship between the condition, performanceand importance is presented in table M2-2 on page 90.

91


92

Module 2

Tab

le M

2-3:

Sum

mar

y o

f st

eps

for

Rap

id A

sset

Ap

pra

isal

(S

ourc

e: B

urto

n 20

00 in

MA

INTA

IN T

hem

atic

Pap

er N

o.8)

Dat

a an

d D

etai

lsP

urp

ose

Sys

tem

ove

rvie

w�

insp

ect

cana

l and

dra

inag

e sy

stem

To g

ain

an o

verv

iew

and

the

“fe

el”

of t

he ir

rigat

ion

�p

relim

inar

y ta

lks

with

farm

ers

and

ser

vice

pro

vid

er s

taff

and

dra

inag

e sy

stem

as

a gu

ide

to t

he a

pp

roac

hon

sys

tem

man

agem

ent,

op

erat

ion

and

mai

nten

ance

to b

e ad

opte

d fo

r th

e as

set

surv

ey.

Not

e th

e ge

nera

l ty

pe

and

con

diti

on o

f the

infr

astr

uctu

re, t

he t

ype,

co

nditi

on a

nd q

ualit

y of

the

farm

ing.

Ob

tain

gen

eral

Bac

kgro

und

dat

a in

clud

es:

Gai

n m

ore

det

aile

d in

sigh

t of

the

sch

eme,

the

layo

ut,

bac

kgro

und

dat

a�

map

sp

rod

uctio

n le

vels

, man

agem

ent,

op

erat

ion

and

�cl

imat

ic d

ata

mai

nten

ance

pro

ced

ures

. Sta

rt t

o fo

rmul

ate

�sc

hem

e cr

opp

ing

pat

tern

initi

al o

pin

ion

of t

he m

anag

emen

t, o

per

atio

n an

d�

aver

age

crop

yie

lds

and

pro

duc

tion

mai

nten

ance

, and

its

imp

licat

ions

for

�sy

stem

op

erat

ion

and

mai

nten

ance

pro

ced

ures

sche

me

per

form

ance

.

Ob

tain

and

pro

cess

�

crop

typ

e, y

ield

s an

d p

rod

uctio

n p

er t

ertia

ry u

nit

Ob

tain

and

pro

cess

dat

a to

qua

ntify

sch

eme

det

aile

d s

yste

m�

irrig

atio

n w

ater

sup

ply

at

prim

ary,

per

form

ance

and

pot

entia

l pro

ble

ms.

per

form

ance

dat

ase

cond

ary

and

ter

tiary

hea

d r

egul

ator

s�

pro

cess

dat

a re

ady

for

anal

ysis

Iden

tify

cur

rent

�

anal

yse

per

form

ance

dat

a to

iden

tify

pro

duc

tion

pro

ble

ms:

A

naly

se d

ata

to q

uant

ify c

urre

nt a

nd p

oten

tial

and

po

tent

ial

ow y

ield

s, d

iffer

entia

l pro

duc

tion

per

form

ance

leve

ls o

f per

form

ance

and

cur

rent

and

per

form

ance

leve

l �

hold

dis

cuss

ion

with

farm

ers

and

ser

vice

pro

vid

er s

taff

des

ired

leve

l of s

ervi

ce p

rovi

sion

. Id

entif

y th

e ga

ps

and

cur

rent

and

to a

scer

tain

vie

ws

and

per

cep

tions

on

curr

ent

and

pot

entia

lan

d c

onst

rain

ts a

nd a

sses

s th

e lik

elih

ood

d

esir

ed le

vel o

fp

erfo

rman

ce le

vel a

nd le

vel o

f ser

vice

pro

visi

onth

at t

he c

ond

ition

and

per

form

ance

serv

ice

pro

visi

on

�id

entif

y an

d q

uant

ify c

urre

nt le

vel o

f per

form

ance

and

leve

l of

of t

he in

fras

truc

ture

mig

ht b

e a

fact

or.

serv

ice

pro

visi

on�

iden

tify

and

qua

ntify

pot

entia

l/des

ired

leve

l of p

erfo

rman

ce a

ndle

vel o

f ser

vice

pro

visi

on�

iden

tify

natu

re a

nd c

ause

of g

aps

and

con

stra

ints

�d

ecid

e if

infr

astr

uctu

re c

ond

ition

and

per

form

ance

may

in

fluen

ce o

vera

ll sy

stem

per

form

ance

93


Ob

tain

mai

nten

ance

�re

curr

ent

mai

nten

ance

bud

get

and

allo

catio

nU

nder

stan

d s

ituat

ion

in r

elat

ion

to m

aint

enan

ce.

dat

a�

annu

al e

xpen

ditu

re o

n m

aint

enan

ce c

ontr

acts

Ans

wer

que

stio

ns: I

s m

aint

enan

ce fu

ndin

g hi

gh,

�an

nual

exp

end

iture

on

emer

genc

y m

aint

enan

cem

oder

ate,

low

? Is

mai

nten

ance

wor

k b

eing

def

erre

d?

�m

aint

enan

ce r

equi

rem

ents

Are

mai

nten

ance

pro

ced

ures

(for

wor

k id

entif

icat

ion/

�

mai

nten

ance

sta

ff, e

qui

pm

ent

and

faci

litie

sp

riorit

isat

ion,

etc

) ad

equa

te?

Is la

ck o

f mai

nten

ance

�

form

op

inio

n on

mai

nten

ance

situ

atio

nan

issu

e?

Det

erm

ine

exte

nt

�ob

tain

ass

et d

atab

ase

Iden

tify

and

cat

egor

ise

asse

t ty

pes

, siz

es, f

unct

ion

etc.

of

asse

t b

ase,

�

obta

in d

esig

n d

raw

ings

(if a

vaila

ble

)P

rep

are

stra

tific

atio

n of

the

sys

tem

and

dec

ide

onst

rati

fy a

nd�

dec

ide

on s

trat

ifica

tion,

sam

ple

set

s an

d s

ize

sam

ple

set

s an

d s

izes

, and

nor

mal

isat

ion

mea

sure

s se

lect

sam

ple

bas

e�

dec

ide

on s

amp

le s

ets

for

each

sam

ple

set

. P

rep

arat

ory

wor

k fo

r ca

rryi

ng

�d

ecid

e on

nor

mal

isat

ion

mea

sure

s fo

r ea

ch s

amp

le s

etou

t as

set

surv

eys.

Car

ry o

ut

For

sele

cted

sam

ple

set

s:O

bta

in t

he fi

eld

ata

for

the

sam

ple

set

s fr

om w

hich

to

asse

t su

rvey

�ca

tego

rise

and

rec

ord

ass

et c

ond

ition

be

able

to

form

ulat

e as

set

cond

ition

and

per

form

ance

�ca

tego

rise

and

rec

ord

ass

et p

erfo

rman

cein

vent

ory

rep

ort.

�ca

tego

rise

and

rec

ord

ass

et im

por

tanc

e

Form

ulat

e as

set

�ex

pan

d s

amp

le s

et d

ata

to fu

ll p

opul

atio

n (”

norm

alis

atio

n”)

Form

ulat

e fin

al fi

ndin

gs o

n th

e in

fluen

ce o

f co

ndit

ion

and

�as

sess

imp

act

of in

fras

truc

ture

on

sche

me

infr

astr

uctu

re c

ond

ition

and

per

form

ance

on

curr

ent

per

form

ance

rep

ort

per

form

ance

and

leve

l of s

ervi

ce p

rovi

sion

and

pot

entia

l/des

ired

sch

eme

per

form

ance

and

leve

l of

�fin

alis

e fin

din

gs o

f inf

luen

ce o

f inf

rast

ruct

ure

on c

urre

nt

serv

ice

pro

visi

on.

Writ

e up

find

ings

as

Ass

et C

ond

ition

sy

stem

and

sch

eme

per

form

ance

and

leve

l of s

ervi

ce p

rovi

sion

.an

d P

erfo

rman

ce R

epor

t.�

Fina

lise

find

ings

on

gap

s b

etw

een

curr

ent

and

pot

entia

l sc

hem

e p

erfo

rman

ce a

nd c

urre

nt a

nd d

esire

d le

vel o

f ser

vice

in

rel

atio

n to

infr

astr

uctu

re c

ond

ition

and

per

form

ance

�

pre

par

e A

sset

Con

diti

on a

nd P

erfo

rman

ce R

epor

t

94

Module 2

Tab

le M

2-4:

Exa

mp

le o

f p

rofo

rma

for

asse

t su

rvey

dat

a co

llect

ion

(Bur

ton

2000

)

Can

al/d

rain

nam

e:

Cha

inag

eA

sset

Fa

cet

Con

diti

onP

erfo

rman

ceIm

por

tanc

eD

isch

arge

Lead

ing

dim

ensi

ons

Rem

arks

Sta

rt-F

inis

hty

pe

Rat

ing

Rat

ing

Rat

ing

cap

acity

(1-5

)(1

-5)

(1-5

)(m

3/s)

Sta

rtFi

nish

Hei

ght

Wid

thLe

ngth

Oth

er(K

m)

(Km

)(m

)(m

)(m

)

Sur

vey

cond

ucte

d b

y:

Nam

e:D

ate:

Po

siti

on:

Annex to Module 2

A distinction needs to be drawn between thecondition of an asset and the impact that condition levelhas on the performance of the asset in its definedfunction. It is possible to find an asset, such as a crossregulator, which is in poor condition but which is stilladequately performing its function. In the UK waterindustry it was found that money was being spent onimproving the condition of assets whilst there was littlevisible or felt improvement in the system’sperformance. With limited availability of funds thefocus has turned towards expenditure on assets tomaintain or enhance performance leading towardsmaintaining or enhancing the level of service provisionto the customer.

Splitting the assessment of the asset into two parts,condition and performance creates difficulties in:

� surveying of the assets� deciding on priorities for expenditure� deciding how performance and condition are linked.

The key to overcoming these difficulties is to be clearand explicit about the function of each asset. In thesections below the procedures are outlined forcondition,performance and importance grading.

Asset condition inventories are now becoming fairlystandard in many civil engineering systems. In somecases, significant steps have been made towardsstandardising the condition grading. For conditiongrading the asset must be divided up into it maincomponent parts, termed “facets”,and the condition ofeach of those parts assessed separately. Thus a gatedcross regulator might be divided into its upstreamwingwalls, upstream base and cutoff, throat section,downstream wingwalls, downstream base and cutoff,and gate.

For condition grading two basic questions need to beborne in mind when surveying the asset:

95

Annex to Module 2

Differentiating

between asset

condition and

performance

Condition grading

of assets

96

Module 2

� Is the asset safe?� Does the asset require repair?

Much condition grading relies on visual observation,though in the U.S. Army Corps of Engineers (USACE)Directorate of Civil Engineering the conditionassessment includes physical tests such as load testing.In the UK water industry a 5-point grading system hasbeen adopted as shown in Table M2-5. In additioncolour photographs illustrating the various conditiongrades have been used to minimise the subjectivityinvolved when assessing asset condition (see e.g.Glennie at al. 1991).

Performance grading seeks to assess the degree towhich the asset is able to perform its function. The

Performance grading

of assets

Table M2-5: Example of standardised condition grading for a concrete bridge over a canal or drain (Source: Burton 2000 in MAINTAIN Thematic Paper No. 8)

Concrete Bridge Structures

ConditionGrade Description

1 No visible defects. No more than hairline cracks, no signs ofany honeycombing or spalling.

2 Wider cracking, greater than 0.5 mm. Localisedhoneycombing and spalling. Concrete flaking. Signs of previous repair.

3 Rust staining. Spalling of concrete or exposure ofreinforcement. Extensive or widespread honeycombing.Evidence of weathering/erosion. Surface covered in vegetation

4 Extensive/widespread concrete spalling. Extensive exposureof reinforcement and rust staining. Signs of reduced structuralintegrity.

5 Clear evidence of structural failure or that failure is imminent.

assessment is for the asset as a whole if it has only onemajor function, or for relevant aspects if it has severalfunctions. The main questions,which need to be bornein mind when carrying out the performance survey,are:

� can the asset perform its function or performanceservice?

� can the asset perform to its design capacity?� how does the performance of the asset influence

system performance?

The performance grading system is similar to that forcondition grading, with five grades. To focus on thefunctionality aspect, a Function Statement is attachedto each asset that defines its function. An example ofperformance grading for a canal head regulator is givenin Table M2-6. The performance grading must relatecarefully to the Function Statement, thus for a headregulator the performance relates to the structure’sability to control the flow entering the canal, whetherit be to the design maximum,or to zero.

A feature of performance grading of the asset is thatit may require testing of the asset. Thus in the case of ahead regulator the gate must be operated during thesurvey to see that it can pass the design discharge, orclose off the supply completely.This can be in conflictwith condition grading which may require the systemto be drained in order to inspect parts of the asset whichare normally submerged.

The importance of an asset is a measure of itsstrategic importance to the overall functioning of theirrigation system. Influencing factors include:

� function� area served downstream� area affected or influenced by structure� cost of replacing the structure� number of people affected by structure� danger to health and safety of asset failure� impact on scheme performance

97

Annex to Module 2

Importance grading

98

Module 2

Table M2-6: Example of standardised performance grading for acanal head regulator(Source: Burton 2000 in MAINTAIN Thematic Paper No.8)

Canal head regulator

Statement of To control and regulate water entering a canal Function or from designmaximum discharge to zero flow.Performance Service

Performance DescriptionGrade

1 The structure can pass the design maximum flow, andcan be shut completely to pass zero flow. There is noseepage around or under the structure into the canal.

2 The structure has restrictions on its ability to pass thedesign maximum flow, cannot be shut completely, and/orthere is seepage around or under the structure into thecanal. Canal discharge is limited to 80 % of design, orthe discharge entering the canal cannot be reducedbelow 20 % of design.

3 The structure has significant restrictions on its ability topass the design maximum flow, cannot be shutcompletely, or there is significant seepage around orunder the structure into the canal. Canal discharge islimited to 60 % of design, or the discharge entering thecanal cannot be reduced below 40 % of design.

4 The structure has severe restrictions on its ability to passthe design maximum flow, cannot be shut completely, orthere is severe seepage around or under the structureinto the canal. Canal discharge is limited to less than 40% of design, or the discharge entering the canal cannotbe reduced below 60 % of design.

5 There is no control of discharge through, around or underthe structure. Discharge entering the canal may be zeroor greater than 100 % of design.

99

Annex to Module 2

There is no consensus yet as to a standardisedapproach to classification of importance. Based on thework of Cornish and Skutsch (1997) and IIS (1995) thefollowing algorithm is proposed to develop anImportance grading for an asset:Importance grading = (ai/A) x FI

Where:ai is the area influenced by the asset. Bridges, roads,

escape structures, etc, are assigned a service areaequal to that of the canal reach on which they occur

A is the total command area of the irrigation schemeFI is the Function Index taken from Table M2-7

The classification of Table M2-7 is somewhatsubjective, for a given scheme or schemes it may beadjusted to suit. Note that ai relates to the areainfluenced by an asset,thus a cross regulator and a headregulator at a secondary canal division point will havethe same importance grading as they both influence thesame total command area.

Table M2-7: Asset Function Index for determination ofimportance grading for irrigation and drainage infrastructure(after Cornish and Skutsch 1997)

Function Index(classes)

5 4 3 2 1

Diversion weir Scour sluice Canal reach Drain reach Inspectionroad

Embankment Cross drainage Head regulator Drop/chute Bridgedam culvert

Intake works Aqueduct Cross regulator Side weir

Pump station Syphon Measuring Tail escapestructure

Barrage Sediment trap

100

Module 3

Module 3

Rapid Assessment of Economic Incentives forMaintenance (REA)(Supporting Document:MAINTAIN Thematic Paper No.11)

When talking about incentives, we generallydifferentiate between material and immaterialincentives. With respect to maintenance, individualfarmers may feel strong obligations toward thecommunity or group to contribute their share andhence gain much of their motivation throughimmaterial – in this case socially-based – incentives.However the individual farmer,as well as a communitygroup or an irrigation agency,will all base their decisionto devote time and inputs to maintenance activitiespredominantly on some kind of economic calculation,asking the question whether or not the maintenanceexercise will be “worth the effort”.2

To find out in detail whether this is the case or not,each actor or group of actors will have to examine boththe costs incurred by contributing inputs tomaintenance, and the benefit stream resulting fromsuch contributions. On the one hand this requiresassessing maintenance needs on the basis of intendedlevels of irrigation service (see MAINTAIN Module No.1).On the other hand it demands knowledge about thebenefit stream generated by the irrigation serviceprovided and the impact that different levels ofmaintenance have on that benefit stream. We havepointed this out repeatedly before.

A classic economic analysis would look at therelationship between incremental expenditures on

Why do we need a

“Rapid Economic

Assessment” (REA) of

Maintenance Needs?

2 Economists would tend to interpret even a socially-based decision asan economic one. They would argue that the farmer would consider thecosts incurred in case of non-participation. Such costs may come aboutas social sanctions, as costs of not being able to claim social solidarityin the future‚ as psychological costs of losing social standing etc.

system maintenance and incremental units ofagricultural output. Nominally, expenditure onmaintenance would be increased until its cost was justequal to the value of an additional unit of output.However, such a relationship involves extremelydifficult-to-measure variables, substantial time lags, anda great many intervening variables.Moreover,the natureof the relationship may change over time. All of thesefactors mean that, in practice, even in large scalesystems, such a traditional economic treatment isseldom a realistic option to assess and decide uponmaintenance needs.

What other economic assessment can the involvedactors undertake that may support decision makingwith respect to maintenance efforts? We believe that allthe contributing actors can do – and in reality are doing– a “rapid economic assessment”of maintenance needs.Such an assessment hence substitutes for a full fledgedeconomic assessment that in most cases will be toosophisticated to be implemented in practice.

The REA proposed here starts from the premise thatevery actor potentially involved in maintenance – i.e.every organization, group or individual – will do arough benefit-cost calculation on his/her own.

On the cost side, an actor will consider theopportunity cost of maintenance inputs or necessarycontributions,i.e.the benefits forgone by spending timeand inputs for maintenance activities. This is animportant consideration for everyone: for farmersduring labor intensive periods of the vegetative cycle;for farmers that have other activities apart fromirrigation; for part-time farmers; for maintenance staff

101

Rapid Assessment of Economic Incentives for Maintenance

What is the rationale

behind the Rapid

Economic

Assessment (REA) of

maintenance needs?

The purpose of this Module is to introduce the basic philosophy behind sucha “rapid economic assessment of maintenance needs”. Doing so, the Moduleintends to draw attention to basic benefit-cost considerations that are doneby different actors and that create the economic incentives to embark or notto embark upon intensive maintenance efforts.

who may use the time for activities that are moreglamorous;for managers of irrigation agencies who maytend to devote more time and scarce funds to activitiesthat are more visible and politically rewarding. All ofthese actors may have different perceptions of thedimension of the opportunity costs incurred,but all ofthem will have such a perception.

They also will have a perception as to the benefits oftheir own inputs or contributions to maintenance.Andthese benefits will accrue differently to different actors.Moreover, we have discussed in chapter 2.3 of thisGuide that maintenance is a typcal “future good”,sincebenefits of maintenance accrue in a sometimes distantfuture.Consequently many actors will heavily discountsuch benefits, especially poor farmers that have tostruggle for survival.The focus therefore will not be somuch on the benefits of maintenance but instead on therisks or the potential losses brought about by not doingmaintenance. The different perceptions of the risk ofdeficient maintenance or of outright neglect togetherwith the perceptions of the opportunity cost ofnecessary maintenance inputs therefore determine theeconomic incentives of actors to contribute heavily,only slightly,or not at all to maintenance efforts.

As to the risks involved in not doing maintenance,Levine (MAINTAIN Thematic Paper No.11) comes upwith an interesting argument.He contends that becausesystems are usually designed with excess conveyancecapacity, and because more effective systemmanagement can compensate for some degree ofsystem deterioration, that loss of system benefits lagsdeterioration by a significant number of years. Thiscertainly will help to lower the perception of risksconsiderably and thus favour the neglect ofmaintenance efforts.

Moreover, from the point of view of local decisionmakers there may even be a certain risk involved indoing maintenance. Rehabilitation financing isgenerally available from national budgets orinternational lenders and donors on concessional

102

Module 3

terms,while maintenance has to be financed from localfunds. Hence there is the risk that, in doingmaintenance,one may forego the external rehabilitationfunds. Consequently, a cyclic pattern of minimalmaintenance and controlled deterioration, followed byexternally-financed3 rehabilitation, is a logical andeffective strategy to follow.

REA tries to assess the economic incentives ofdifferent actors – either by guessing, questioning or inthe process of workshop discussions – by orderingthese incentives into four major clusters.These clustersare formed by a matrix in which the perceivedopportunity costs of the inputs needed for maintenancecomprise one axis and the perceived risks to deficientmaintenance are shown on the other.An assessment ofthese factors will lead to a positioning in one of the fourquadrants, represented in Fig.M3-1 on page 106.

Quadrant 1,in which the perceived opportunity costof maintenance inputs is high,but the perceived risk ofconsequences to limited maintenance is low,representsthe assessment of farmers in many circumstances,particularly in low-income developing countries. Insuch cases, farmers are likely growing small grains,e.g.rice, maize, wheat, partly for own consumption andpartly for sale. Market value of these crops is relativelylow, and at moderate input levels, they are onlymoderately sensitive to water stress.At the same time,resources for all expentitures are generally scarce,particularly for farmers who have to tend to otherfarming activities apart from irrigation.

The situation in Quadrant 1 also reflects a commonperception of agency staff with respect to maintenance.In most circumstances, staff members cannot be heldaccountable for maintenance deficiencies,and normallyno premiums are paid for good maintenance service.Moreover,since budget allocations to the agency are nottied to the quality of service to the water users, from

103


How to go about a

Rapid Economic

Assessment of

maintenance needs?

3 External to the irrigation system.

their point of view the risks related to poormaintenance are low. Moreover, they may have othermore attractive activities to do – design work forconstruction, for example – so that perceivedopportunity costs of spending time on maintenanceactivities are high. No wonder then, that these actorswill perceive few economic incentives to engage formaintenance. They will tend to just do the minimumonly or neglect maintenance altogether.

In Quadrant 2, both the opportunity cost of inputsand the risks related to system non-performance arehigh.This is a situation which, when we look from thepoint of view of the farmers, is more likely to be foundin middle and higher income countries, where highervalue crops are cultivated under high-input regimes,andproduction is almost entirely market-oriented.Expectations related to benefits from good systemperformance are likely higher than in Quadrant 1.Thisleads to higher incentives to ensure reasonable systemmaintenance. However, since opportunity costs arehigh – e.g. through attractive off-farm employment –efforts will generally be targeted on insuring a highstandard of reliability for critical system facilities only.

From the point of view of agency staff members,Quadrant 2 may correspond to the maintenance ofcritical pieces of infrastructure. Critical infrastructure,in case of failure, may cause serious damage, loss ofincome or even life and thus would clearly rebound tothe maintenance staff. This generally is the case withdam structures or with major structures of the mainconveyance system. This is why, despite highopportunity costs for maintenance inputs, agency staffmay perceive sufficient incentives to engage inpreventive maintenance efforts related to thesestructures.

In Quadrant 3, the actors perceive low risks as wellas low opportunity costs. For the farmers such asituation may arise during periods of low agriculturalactivities or in circumstances when maintenance worksare heavily subsidized. Substantial contributions of

104

Module 3

external funds may provide economic incentives tofarmers to embark on maintenance activities whichthey would otherwise neglect.

Similar circumstances may hold for the staff ofirrigation agencies. In cases where maintenance jobsare the only employment possibilities at hand,workersmay tend even to maintenance activities that have onlylittle or no influence on system performance, e.g.painting rails, cutting grass at the side of access roadsetc.

In Quadrant 4, the risks related to incidents ofsystem non-performance are high, while opportunitycosts of providing maintenance are low.

From the farmers point of view this may be the casein high-commercial farming, where even small failuresor interruptions of the intended service level can causelarge losses of income. This may, for example, be thecase in modern green house farming.Moreover, if laborcosts for maintenance are cheap, for example due toready availability of migrant labor, the option of“maximum maintenance”may be the appropriate one.

From the view point of agency staff, such a situationmight arise in circumstances where the staff isresponsible for maintenance in a high commercialirrigation environment – as e.g. in the Central Valley inCalifornia (see MAINTAIN Case Study No.4).Here,highexpenditure for maintenance may be seen as proof ofthe importance of the maintenance service and evenhelp to expand staff numbers.Hence opportunity cost– in the eyes of the staff members – may be extremelylow. No wonder then, that in such circumstances atendency to do “gold plated”maintenance may prevail.

The important fact to keep in mind when going abouta REA, is that this is not an objective economicassessment but an exercise that tries to assertain thesubjective perceptions of particular actors. These areperceptions, however, that form the basis for actualbehavior. This means that different actors may havecompletely different perceptions about economicincentives to do maintenance. Even when the top

105


106

Module 3

management of an irrigation agency tries to quantify“objectively” the risks involved in deficientmaintenance and attempts to assess the opportunitycosts involved when doing maintenance, farmers,agency staff or other actors may perceive the situationin a completely different manner. An analysis of actorspecific incentives may further clarify why suchdeviations come about (see Module 10).

Fig. M3-1: Matrix for Rapid Economic Assessment of

maintenance needs

Perceived

opportunity

cost of

maintenance

Module 4

“Service Interaction Analysis” (SIA)(Supporting Documents:MAINTAIN Case studies No.1to 6; MAINTAIN Thematic Paper No. 7, 10 and 11 andGTZ publication No.263)

Throughout this Guide,the term provision or serviceprovision refers not only to primary irrigation servicessuch as water conveyance and delivery and secondaryservices such as maintenance. It also refers to“supporting services”such as information and adminis-trative services, coordination and representation.

To analyse the provision of all primary,secondary andsupporting services in a service delivery system meansto find answers to the following questions:

� What are the primary, secondary and supportingservices provided?

� Who are the “consumers” or beneficiaries of theseservices?

� Who are the providers of these services?� Who are those who pay for these services or in some

other way provide a “return”?� Who arranges for and monitors the delivery (or

“production”) of these services, and what are thesearrangements?

� What arrangements are made for the financing ofservice provision?

107

Module 4

This Module provides a practical way for how to answer the first four of thesequestions. Answers to the fifth question are addressed in Module 5, answersto the last question are found in Module 9. This Module 4 may be useful in allof the basic institutional contexts described in chapter 4. However, it will beparticulary helpful in situation B, where “Institutional Change Approaches” arepursued.

Answers to the first four of the above questions seemto be obvious in some cases. However, answers aredifficult to find when service provision involves manydifferent actors (a network of stakeholders) and whenvarious supporting services are needed to make theprovision of primary and secondary services functional.This is the case in most processes of maintenanceprovision:the provision may be arranged,paid for,usedand provided by different actors and it may require arange of supporting services, such as provision ofinputs, information, monitoring and auditing services,official representation of interests (viz., of farmers, et.al.), etc. In such a complex web of relations,“ServiceInteraction Analysis” (SIA) can quickly bring abouttransparency. SIA is a set of tools designed to helpidentify systematically the services that are being orshould be provided.The tools can also make it easier toanalyse the problems associated with the provision ofservices in a way that takes into account thedistinguishing features of services.

For example,SIA can serve as an instrument to clarifymutual expectations of the provider and the client inthe provision process. As such it will be a first step intailoring mutual agreements or contracts with respectto the details and conditions of the provision of servicesin question.

Below, we present a sequence of steps that can befollowed when performing a SIA exercise.However,thissequence can be handled with flexibility,depending onthe particular problems at hand.

The SIA might be applied in settings in which theprimary or supporting services are vague and not welldefined and where the service interactions betweendifferent actors are unclear or problematic.

The SIA can be targeted at the maintenance situationof the whole irrigation scheme or at the maintenanceprovision of a particular section of the hydraulicinfrastructure (see the discussion below on the so-called “hydro-institutional service chain”). In many

108

Module 4

What is SIA and what

is its purpose?

How to organize

a SIA?

109

Service Interaction Analysis (SIA)

cases it might be advisable to perform a SIA exerciseboth for the primary service of water delivery and thesecondary service of maintenance provision.

The ideal way to apply the SIA is within a workshopsituation in which the most important role playersparticipate.The central problem to be addressed by theworkshop has to be identified beforehand togetherwith the initiators of the workshop. This discussiongenerally reveals who the respective participantsshould be.

In full, the SIA can embrace the following steps:

� Identify and visualize the tapestry of relations and the“hydro-institutional service chain”(see below);

� Identify the major services or support services, thateach of the involved organizations or units, i.e. each“actor”, is supposed to provide to the other actors.To do this, establish a “Service Interaction Program”(see below);

� Discuss major problem areas;� Choose one particular organization or unit as the

focal point of a more in-depth analysis and then,related to this organization/unit answer the followingquestions:– What is the range of services to be provided by this

actor and to whom are these services supposed tobe provided? (i.e.,the “Service Provision Program”of the actor in question),

– What is the range of services supposed to bereceived by this actor and by whom are theseservices supposed to be provided? (“ServiceReception Program”of the actor in question);

� Analyse the strengths and weaknesses of theseprovision and reception programs, including toidentify gaps between actual and intendedprovisions;

� If needed, analyse and assess in the same way theinternal interactions within the focal organization,perceiving each internal unit of this organization asthe provider and receiver of certain internal services.

In the following we give short outlines to each ofthese steps.

The first step aims to render transparent the complextapestry of relations that exists between organizationsworking within a service network.For a specific serviceor support service, such as water delivery ormaintenance, all the major stakeholders are to beidentified.In other words,all the actors that are involvedin any of the functions of providing, arranging, payingfor, regulating, monitoring, providing support for aparticular service are identified and listed.

Table 3 in chapter 2.4 indicates how numerous theinterrelated actors in the field of maintenance can be.It lists all the different organizations,entities and groupswith whom the Nienburg/Weser MaintenanceAssociation in Germany has working relationships inorder to accomplish its purposes.

Before carrying out this part of the SIA one needs tobe clear about the scope of the exercise. Do we wantto analyse service interactions throughout the wholesystem or should we rather focus on parts of the systemonly? This question may be answered more readily afterdrawing up a so-called “hydro-institutional servicechain.”With such a service chain one follows the flowpath of the water from the source or the storage facilityto the water delivery at the farm gates.Along this paththe service chain identifies the parts of the system thatprovide different supporting services, that require dif-ferent maintenance efforts and that involve a differentset of actors.Table M4-1 shows the example of a hydro-institutional service chain as it corresponds to waterconveyance and delivery as well as to maintenance inthe Neste System in Southern France. This system isdescribed in detail in MAINTAIN Case Study No.2.Thetable also indicates the major actors involved.

Once the scope of the exercise is clear, it often helpsin a workshop setting to draw a sketch of the “servicenetwork”that will be analysed more in detail.This initialstep of the SIA helps to draw the attention of workshop

110

Module 4

Identifying the

tapestry of actors

and the “Hydro-

Institutional Service

Chain”

111


participants to the complex web of relations that existsand that needs to function in order to render theintended service provision effective. In mostcircumstances it is sufficient to draw a rough sketch ofthe service network and not indicate each and everyinterface.

Table M4-1: Example of a Hydro-Institutional Service Chain (The Neste System in Southern France, described in MAINTAIN Case Study 2)

Infrastructure Services provided Providers of Arrangers, payers, usersthe services and regulators of services

Storage dams � Water storage and � “Electricité de � State, EdF, irrigators,in the Pyrenees power generation France” (EdF) public and private users

� Maintenance of dams with private of electricityand power generation contractors

Neste Canal � Water conveyance and � Compagnie � State, Public, CACGsupply to 17 small d’ Aménagement

rivers and streams des Coteaux� Maintaining all infra- de Gascogne

structure in the Neste (CACG)canal system

Small rivers � Ensuring minimum � CACG � Agence de l’eauand streams flow rates in � Public

rivers and streams � Ecology� Specific users (e.g.fisheries)

� Supplying specific � CACG � Irrigators in concession amounts of water into perimetersthe water courses � Water user associations

� Domestic water supplycompanies and villages

� Comité Neste� State entity DDAF

as »police de l’eau«

� Maintaining river banks � Littoral � Littoral landownerslandowners � State entity DDAF

as »police de l’eau«

Franchise � Providing set quantities � CACG � Individual farmers in the Irrigation of water to the franchise systemsPerimeters perimeters (FP’s) during � Comité Neste(“périmètres en a ten year period with � State (DDAF)concession”) one year extensions � Conseil Administratif

� Maintain the hydraulic infrastructure of the FP’s

112

Module 4

An important point here is the fact that a servicenetwork can also be drawn up for relations betweendepartments, divisions and other entities within anorganization. In this case, one assumes that goodfunctioning of the organization requires effectiveinternal relationships that may be thought of as internalservice exchanges.

Figure M4-1 shows the set of actors involved inprovision of maintenance services for the primary andsecondary canal infrastructure of the “Lower SeyhanIrrigation Scheme”in Turkey.

A second and often quite helpful step in a SIAexercise is the elaboration of a matrix of servicerelations.The procedure is simple when representativesof the various stakeholders are present in theworkshop.Each group of stakeholders is asked to define

Establishing a matrix

of service relations

Figure M4-1: Field of actors involved in maintenance service

provision – Lower Seyhan Irrigation Scheme, Turkey (Source:

Scheumann and Vallentin, 1999 in MAINTAIN Case Study

No.1)

a small number of the most important services that theyare supposed to provide to anyone of the other groups.

This exercise results in a matrix where everystakeholder group is listed in a horizontal row and in avertical column.Such a matrix allows a quick overviewof the most important services and theinterdependencies within the service network.Thematrix shows the services to be provided from onestakeholder group (listed in the row) to another group(listed in the column). Normally, representatives of therecipients of these services will be present in the SIAworkshop. This enables participants to identifydeficiencies in the actual provision of services.

One of the central activities of the SIA is to make anoverview of the entire range of services provided byone of the organizations or entities identified above.This illustrates the full breadth of the services providedby this actor. To this end a list is drawn up of all theservices provided by the organization and of therecipients of these services. The list is then brokendown and categorized by type of service and recipient.This tends to reveal a number of services that wouldotherwise often be glossed over because of theirintangible nature. These can be coordination inputs(liaising, clarifying legal issues, etc.), informationservices (providing specific data,advisory services,etc.)or others.The listing of the whole range of services alsoreveals the entire spectrum of recipients of services –in other words:exchange partners – who have relationsto the organization in question.

Table M4-2 gives an overview over the servicesprovided by German Maintenance Associations .

An important step in a SIA exercise is theidentification of all supporting services provided to aparticular organization. Here again, experience showsthat normally there is a larger number of servicessupplied by various providers than was expected byparticipants before the exercise started. One needs to

113


Defining the range of

services provided by

a particular entity or

organization

Defining the range of

services provided to a

particular entity or

organization

114

Module 4

Table M4-2: Range of services provided by German MaintenanceAssociations in general and the Nienburg Association in particular.

(Source: Huppert and Urban,1998 in GTZ publication No. 263)

Service pro- Members Nonmembersvided to �

Type of Permit- Other Social service issuing institutions environment� authorities

1. Active � Clearance � Ecologymaintenance � Repair (clearance, services � Maintenance repair,

maintenance)

2. Internal � Planning/services engineering

services� Internal admini-

stration services� Administration of

membership fees� Updating records

of land use

3. Coordination � Internal � External � Externalservices information coordination coordination

(obtaining (liaising andnecessary coordinatingpermits) with other

bodies)� Clarifying

legal issues and cases

4. Information � Information � Trade fairs/ Services events exhibitions

� Informationevents

be aware that deficiencies in any of these servicerelationships may be a cause for sub-optimalperformance of the overall service provision.

As an example, the maintenance of a drainage canalsystem may be obstructed because some littoral farmsinhibit maintenance works. Also, they do not provide

115


* D

enot

es p

rob

lem

s ar

eas.

Tab

le M

4-3:

Ran

ge

of

serv

ices

pro

vid

ed t

oG

erm

an M

aint

enan

ce A

sso

ciat

ions

in g

ener

al a

nd t

oth

eN

ienb

urg

/Wes

er A

sso

ciat

ion

in p

arti

cula

r (S

ourc

e: H

upp

ert

and

Urb

an, 1

996)

the responsible maintenance provider with informationabout specific problems they face with themaintenance program. Unless the relationship withthese actors can be made functional,such “obstructors”can seriously hinder the overall program.

Table. M4-3 demonstrates the large number ofrelationships that must be managed by GermanMaintenance Associations.

Analysis of the strengths and weaknesses in serviceprovision programs enables the identification ofproblems or difficult operations among the largenumber of supporting services.

To assess the quality of the services provided,workshop participants are asked to rank the serviceslisted on a scale from 1 to 5, from very weak to verystrong. Experiences with SIA show that ranking tasksmakes decision makers more aware of the so-called“soft” services, such as provision of data and otherinformation, maintenance of informal communicationchannels,marketing activities and so on.

Figure M4-2 presents the results of a workshop,where participants were asked to assess the strengthsand weaknesses of services provided by a particularorganization (The Nienburg Maintenance Association)and the strengths and weaknesses of services providedby the entire water infrastructure maintenancesubsector (the German Maintenance Associations).

Services or supporting services that have beenranked unsatisfactory in the before mentioned exerciseare further scrutinized. Now, the special nature ofservices must be taken into account. Various kinds ofservices, especially the so-called “interpersonalservices”such as consultancies,can only be provided inclose collaboration with clients. This means that theproblem of providing services must be seen not onlyfrom the viewpoint of the provider, but also from theviewpoint of the recipient and at the interface betweenthe two sides where the interaction takes place.Thus,

116

Module 4

Analysing strengths

and weaknesses of

service provision

Analysing problems

with service

interaction

117


Figure M4-2: Services provided by the Nienburg/Weser MaintenanceAssociation: Analysis of strengths and weaknesses.(Source: Huppert and Urban, 1998)

Services very weak weak moderate strong very strong

Clearance � �

Maintenance � �

Care of wood � �

Planning � �

Engineering services � �

Internal administration � �

Administration � �of membership fees

Updating records � �of land use

Internal coordination � �

External coordination- obtaining necessary � �

permits

Liasing and coordination with other bodies � �

Clarifying legal issues and causes � �

Information events � �

Issuing circulars � �

Trade fairs/exhibitions � �

Ecology � �

Ecological engineering services � �

Note: � Associations in general.� Nienburg Association.

an analysis of service interactions has to look atpotential problems identified at three separate levels:

� Problems involved in providing the services,� Problems involved in receiving the services,� Problems during the interaction between provider

and recipient.

118

Module 4

119

Analysis of Property Rights and Authority Systems

Module 5

Analysis of Property Rights and AuthoritySystems(Supporting Documents: MAINTAIN Thematic PapersNo.6 and 10).

In Chapter 2.4 of Part One of this Guide we have seenthat processes of infrastructure service provisionrequire certain functions to be assumed by the involvedactors:these are the functions of arranging the service,providing the service, paying for the service andconsuming or using it.

Property rights and authority systems define who isentitled or obligated to assume these various functions.Hence,they establish the roles of the actors in exchangerelationships.They constitute the backbone of servicerelationships in irrigation service provision, since theydefine who is entitled or not entitled to act in a certainway, who can reap benefits, and who must bear coststhat result from the exchange. Beyond conveyingauthority and assigning responsibilities, however,property rights and authority delegation also play a vitalrole in creating incentives for the various actors toperform particular actions. Hence, to trace strengthsand deficiencies in the system of property rights andfunctional authority means at the same time to exploreincentive deficiencies in the provision system at hand.

Why are property

rights and the

authority to perform

service functions

important topics in

the context of

maintenance service

provision?

The purpose of this Module is to give practical advice on how to detectdeficiencies in property rights systems and authority systems for servicefunctions related to infrastructure service provision in irrigation (with particularemphasis on maintenance). This Module is especially relevant for institutionalcontexts of type B, as defined in Chapter 4 of this Guide. Here, institutionalchange processes need to be initiated. Such change processes often involvechanges in property rights and authority systems.

Property rights to assets or resources can be definedas “the capacity to call upon the collective to standbehind one’s claim to a benefit stream”that results fromthe use of these very assets or resources (Broomley,1991,cited in MAINTAIN Thematic Paper No.6).In mostcases in industrialized societies, the institution backingthe claim is the state (or statutory) legal system.However, this is not the only source of property rights,especially in the case of water and irrigation systemassets. In addition to statutory law, most societies havedevised varying forms of rights and rules pertaining tothe use of water. Local norms and accepted practicesmay differ from statutory law, while irrigation projectregulations may provide yet another basis for propertyrights.Thus, customary and religious institutions, localsociety, or even irrigation projects may be the backinginstitution. These different sources of rights may becontradictory, adding to the complexity of propertyrights,but also allowing for dynamic change.

The above definition of property rights implies thatall property rights involve relationships among people– the holder of the right,those who recognize that right,and those who are backing that right. The reciprocalside of a property right is generally some form of dutyin the context of the provision of the resource. Ananalysis of property rights therefore needs to assess therights and duties involved, as well as the relationshipsbetween the concerned actors.

Authority systems determine the roles and functionsof actors in contexts where multiple stakeholdersinteract.They are often based on property rights: thosewho control the assets or resources hold the authority.However, many authority systems are not based onproperty rights but on some other form of entitlement(e.g. an appointment, law, contract, tradition, orelection).

In service provision, authority systems define theauthority to perform certain functions in the deliveryprocess, i.e. to arrange, provide, consume, and pay for

120

Module 5

What are property

rights?

What are authority

systems?

121


Table M5-1: Authority system for Andhra Pradesh irrigation managementreforms(Source: Svendsen and Huppert, 2000 in MAINTAIN Case Study No.5)

Actor Existence Authority to: Basis of Authorityby Authority of:

ICADD GOAP � Allocate in-state surface water � Indian Constitution� Operate and maintain dams and other � AP Constitution

major structures � AP statutes� Plan and implement irrigation system � Farmer Mgt of

rehabilitation programs Irrigation Systems � Assume control of farmer organizations Act of 1997 (FMISA)

(FO) in event of failure to perform � Krishna Water � Allocate interstate water Disputes Tribunal

Award of 1973

Farmer GOAP � Function as body corporate � FMISAOrganizations/ � Plan and implement water distribution � Changes in Managing � Plan and implement maintenance Revenue CodeCommittees � Maintain landholder register

� Prepare and maintain inventory of system facilities

� Resolve disputes among water users � Levy and collect irrigation fees and

other revenues and maintain accounts� Levy fines for infractions � Issue instructions to CADD Competent

Authority for certain O&M activities� Conduct general body meetings� Elect officers of higher level FOs

Farmer GOAP � Elect WUA president and managing � Land ownership orMembers of committee reps tenancy agreementWUA General � Recall and replace WUA president and � FMISABody managing committee reps

District GOAP � Delineate water user areas � FMISACollectors � Establish WUAs and higher level FOs

with compulsory membership of landholders

� Organize and oversee FO elections� Collect irrigation fees � GOAP Statute

Courts GOAP � Enforce irrigation-related fines � AP Constitutionand penalties � FMISA

Chartered GOAP � Audit FO accounts � StatuteAccountants

Central Water GOI � Regulate dam safety � GOI StatuteCommission

AP = Andhra Pradesh; CADD = Command Area Development DepartmentGOAP = Government of Andhra Pradesh; GOI = Government of India

the service and to establish relationships.Theserelationships may be those linking the actors whoperform these functions.They may also be thosebetween these actors and the people or bodies whostand behind the various claims. Authority systemsdefine both the actors that hold the authority toperform a certain service function and the institutionsthat are the basis of that authority.

Table 5-1 shows an example for an authority system.It depicts a sketch of the authority system that is relevantfor actual irrigation management reforms in AndhraPradesh, India (see MAINTAIN Case Study No.5).

Schlager and Ostrom (1992 cited in MAINTAINThematic Paper No. 6) disaggregate the bundles ofproperty rights into:

� use rights, including access (to enter the resourcedomain,e.g. the right to cross a piece of land,go intoa forest or canal) and withdrawal (to removesomething,e.g. to take water, fodder,or fish); and

� control rights, including management (to modify ortransform the resource, e.g. by planting trees orshrubs, enlarging a canal, or restricting what can beharvested), exclusion (to determine who else mayuse the resource), and alienation (to transfer rightsto others, either by inheritance, sale,or gift).

A more in-depth discussion on the different types ofproperty rights is presented by Meinzen-Dick, 2000, inMAINTAIN Thematic Paper No.6.

When we discuss the authority bases for theperformance of service functions, we need to remindourselves of the different functions that have to beperformed in service provision. These are the abovementioned functions (see chapter 2.4):

� arrange the service� pay for the service� consume/use the service� provide the service

122

Module 5

What are different

types of property

rights to assets and

resources and

different authority

bases for service

functions?

123


Table M5-2 lists in detail the different types ofproperty rights and the corresponding functions androles in service provision.

Analysing systems of property rights and authoritysystems, one needs to check the occurence of thefollowing potential deficiencies:(a) The required rights/authorities are non-existent(b) The required rights/authorities are insufficiently

specified or formalized and are not transparent to allthe involved stakeholders

(c) There are specific insecurities of tenure related tothe required right/authority, e.g.(c1) Non-excludability of other claimants

What are potential

deficiencies of

property rights and of

authority systems?

Table M5-2: Categories of rights, duties and service functions

Type of Detailed rights/duties related Type of Detailed type of authority/dutyproperty to assets/resources service to perform certain functions inright/duty function the service delivery process

Use rights � Access rights Consumer � Authority/entitlement to enter� Withdrawal rights function the service provision process� Right to derive benefits � Right to derive benefits

from the resource from the service provision(“consumer function”)

Control � Right to “manage” the Arranging � Authority to assume therights (1) resource (modify, transform, function “arranging function” for

change the resource) service delivery (determine� Right to exclusion (determine type of service, select

who may use the resource) provider, assign service contract, monitor provision)

Control � Right to alienate the resource Providing � Duty to perform therights (2) (transfer rights to others) function service provision= Transfer rights

Compen- � Duty to compensate/ pay for Payment � Duty to compensate/ paysation duty transfer of rights function for service provision

(c2) The duration of the right/authority is toolimited to create real incentives for theright/authority holders to embark on effectiveand efficient resource use or performance ofservice functions

(c3) The “assurance”of the rights/authorities is toolow,i.e.the enforcability is too weak and hencethe degree to which the right can be defendedis too low.

Deficiencies like these may be due to insufficientcoordination mechanisms (see Module 6) i.e. insuffi-ciently specified contracts, agreements, processes etc.and/or to deficiencies of the institutional framework.

The following steps are essential:

Step1:“Unbundle” the infrastructure services relevant

under the circumstances in question (to do this refer tochapter 2.1 and to Module No. 4 of this Guide) anddetermine what assets/resources and what serviceshave to be looked at.

Normally the following assets/resources and thefollowing related services will need to be considered:

Assets/Resources� Infrastructure, i.e. specific infrastructural elements

(e.g. dams, main system, secondary system, tertiaryand quarternary system)

� Water� Ancillary resources, i.e.land, trees, fish etc.;

(maintenance deficits may have consequences forrights to land,trees and other resources,though theseinterdependencies are seldom considered at present)

Infrastructure services� Provision of the infrastructure � Maintenance as a secondary service to infrastructure

provision� Services of

– water acquisition

124

Module 5

What are important

steps of an analysis

of property rights and

authority systems for

maintenance

service provision in

irrigation ?

– water conveyance– water distribution– water delivery

� Maintenance as a secondary service to wateracquisition,conveyance,distribution and delivery

Step 2:With reference to Table M5-2 above,determine who

holds what type of property righta) in relation to the infrastructural assetsb)in relation to the water resource

Now check on potential deficiencies,using the abovelist.

Step 3:Determine the essential actors involved in the service

delivery system.Fill in a table,corresponding to Table M5-1 identifying

for every actor� which authority stands behind the existence of this

unit (“Existence by authority of…”)� the respective roles and functions (“Authority to…”)� the institutions that back up these roles and functions

(“Basis of Authority”)Now check on potential deficiencies,using the above

mentioned list a-c.

Step 4:Organize working sessions to discuss the

consequences of the identified deficiencies on theincentives of the involved actors. Discuss potentialremedies for these deficiencies.Envisage overall systemchanges, in case such remedies cannot be found.

125


Module 6

Analysing and Improving the Governance ofMaintenance Provision

(Supporting Documents: MAINTAIN Thematic PapersNo.7 and 10; MAINTAIN Case Studies No. 1 to 7; andaccompanying GTZ publication No.263)

Every type of human interaction can be described asan explicit and/or implicit contractual relationship.Theprovisions of the contract specify the mutual claims andobligations in a relationship that must be, in the end,beneficial to both parties. If the process of fulfillingthese claims and obligations is designed in a verifiableand enforceable manner, – in other words, iftransparency and accountability are ensured – bothparties will contribute and receive whatever it takes tomake the joint business a success.The means to achievethis are appropriate “governance modes” and“coordination mechanisms”.

“Governance modes” or “governance regimes” arethought of here as the overarching institutionalarrangements in which a particular contractualarrangement is embedded. For example, a contractualrelationship may be established in the context of amarket system as is the case when irrigationmaintenance is contracted out to a private firm. Or itmay be designed on the basis of hierarchicalgovernance modes,e.g.when an irrigation organization

126

Module 6

What are “governance

modes” and

“coordination

mechanisms”?

The purpose of this Module is to present and discuss various governancemodes and coordination mechanisms that may be relevant in maintenanceprovision. Moreover the Module introduces essential steps of a governanceanalysis of service provision. The final section describes situation specificapproaches to solving governance problems of maintenance provision.

127

Analysing and Improving the Governance of Maintenance Provision

Tab

le M

6-1:

Maj

or

Go

vern

ance

Mo

des

and

Co

ord

inat

ion

Mec

hani

sms

Go

vern

ance

Mo

de

Co

ord

inat

ion

Mec

hani

sms

Enf

orc

ing

Inst

itut

ions

�M

arke

tP

rices

, agr

eem

ents

, con

trac

ts b

ut a

lso:

Cou

rts

/ le

gal s

yste

m, s

tate

(pol

ice)

but

�

law

s, r

egul

atio

nsal

so:

�co

mm

on p

ract

ices

�P

rod

ucer

/ c

onsu

mer

asso

ciat

ions

, uni

ons,

etc

.

�A

dm

inis

trat

ive

Sys

tem

/In

tern

al r

ules

and

reg

ulat

ions

, em

plo

yee

�“S

uper

ior”

Hie

rarc

hy/B

urea

ucra

cyco

ntra

cts,

pla

ns, d

irect

ives

, con

trol

but

als

o:�

Top

Man

agm

ent

�la

ws

�S

uper

viso

ry B

oard

�ex

tern

al a

udits

, dec

isio

ns o

f sup

ervi

sory

�

Cou

rts

/ le

gal s

yste

mb

oard

s /

com

mitt

ees,

etc

.

�C

olle

ctiv

e d

ecis

ion

�M

emb

ersh

ip, v

otin

g, n

omin

atio

n,�

San

ctio

ning

pow

er o

f offi

cial

org

ans

mak

ing

bas

ed o

n el

ectio

ns/

rep

rese

ntat

ion,

form

ally

ad

opte

d r

ules

and

(boa

rd, g

ener

al a

ssem

bly

etc

.)re

pre

sent

atio

nsre

gula

tions

but

als

o:�

Cou

rts

/ le

gal s

yste

m�

Sol

idar

ity

�C

olle

ctiv

e d

ecis

ion

�Tr

ust,

rec

ipro

city

, sol

idar

ity, b

ut a

lso:

�C

olle

ctiv

e/gr

oup

mak

ing

in p

rimar

y gr

oup

s�

per

sona

l inf

luen

ce (b

ased

on

res

ourc

es,

pow

er)

�cu

stom

s /

trad

ition

etc

.

�C

olle

ctiv

e b

arga

inin

g�

Neg

otia

tions

, agr

eem

ents

but

als

o:�

Pub

lic o

pin

ion

(e.g

. med

ia, e

tc.)

�ex

tern

al a

rbite

r

�C

harit

y S

ytem

s�

Don

atio

ns, r

aise

d fu

nds,

but

als

o:

�M

onito

ring

by

don

or�

Pre

cond

ition

s, q

ualif

icat

ion

syst

ems

delegates maintenance responsibilities to a specialmaintenance department within the same organization.Or it may be arranged within the framework of a systemof mutual help and reciprocity, as is the case in manyfarmer managed irrigation systems. Table M6-1 lists anumber of major governance modes.

“Coordination mechanisms” – also termed“contractual governance” or simply “serviceagreements” – are the particular agreements, rules,contracts,regulations and common practices that makea service relationship function. Coordinationmechanisms are embedded in overarching governancemodes. Hence different governance modes allow for arange of particular coordination mechanisms, as isindicated in Table M6-1. For example, in a marketrelationship,the price may be an important mechanismthat helps govern the relationship.In a hierarchy,plans,programs and directives are dominant mechanisms.Table M6-1 lists some of the common coordinationmechanisms,the governance modes they are related to,and institutions that may help enforce compliance withthe agreed upon terms of the service relationship.

As we have said:governance modes and coordinationmechanisms are the “backbone” of institutionalarrangements for provision systems of any kind. Inpractical cases, often several of the above listedgovernance modes combine to form the institutionalarrangement for the provision in question.

When a group of labourers provides maintenance toa private entity, e.g. a private irrigating farm(corresponding to the middle left quadrant in Table M6-2 below),two governance modes have to function.First,a private contract needs to be concluded between thegroup and the private farmer. Here, the governancemode of the market will play a role. Second, the groupitself needs to establish internal rules and agreementsas to the respective rights and obligations of each groupmember and how to share the work and the resultingbenefit. Here, the governance mode of collective

128

Module 6

Institutional

arrangements for

maintenance

provision

129


decision making in a small group (a “primary group”)is essential. Coordination of the provision process willonly be effective if both of these modes functionwithout problems.

Another example: If maintenance is provided by agovernment entity to a water user association on thebasis of “government vending” (lower left quadrant inTable M6-2 below),then three major governance modesare involved. First, the internal collective decisionmaking of the association has to function in order tobe able to specify the terms of the vending arrangementwith the government.Second,internal decision makingwithin the hierarchy of the government entity needs tobe effective to ensure maintenance provision andcompliance with the contract. And third, thegovernance mode of the market plays a role since it willbe market conditions that determine the terms andconditions of the vending contract.

Maintenance can be provided by private, collectiveor governmental entities as a secondary service to the

Table M6-2: Institutional Arrangements for Maintenance Provision

Infrastructure Service Provided

privatly collectively by government

privatly � Market provision � Market provision � Market provisonof maintenance of maintenance of maintenance to private user to collective entity to government

collectively � Collective contract � Internal maintenance � Collective contract for maintenance provision by WUA for maintenanceprovision to provision to private user government

� Neighbourhood help

by � Government � Free government � Internalgoverment management provision of maintenance

� Government maintenance provision by vending � Government vending government

agency

Mai

nten

ance

pro

vid

ed

infrastructure services described in Chapter 2.1(infrastructure provision and water delivery). Such aprovision can take place in the form of differentinstitutional arrangements. Major arrangements of thiskind are listed in Table M6-2.

When we enter into the details of such institutionalarrangements, we will see that in many cases theyinvolve an even more sophisticated structure ofgovernance modes and coordination mechanisms. Inorder to deal with such complex structures,it is helpfulto visualize the whole provision system in the form ofspecial flow diagrams.

This Guide uses a way of graphical presentation thathas been applied throughout the MAINTAIN CaseStudies. Following the approach of Herder-Dorneich(1986), these graphs indicate the following details (seeFigure M6-1 on page 132):

� The actors concerned (organizations, groups andindividuals) are presented as circles or ovals(sometimes also as squares, see example below).

� Services and supporting services are symbolized byan arrow, with a letter that indicates the kind ofservice in question

� Returns, especially financial returns, are alsoindicated by an arrow in the direction of the flow offinances

� Coordination mechanisms between two actors, i.e.the way in which service provision is governed, isdepicted using a straight line plus a rhomboid shapecontaining either a number that is explained in thetext or letters that represent the mechanism.

Fig.M6-1 shows a typical flow-diagram that presentssuch details for water delivery and maintenanceservices provided by the French “Compagnied’Aménagement des Coteaux de Gascogne”(CACG) toso-called “franchise perimeters” in the region of theRiver Neste. This service system and its majorcoordination mechanisms are described in Box M6-1.

130

Module 6

Visualizing the

governance of service

relationships

131


Box M6-1: Services, returns and coordination mechanisms – theexample of franchise perimeters in Southern France.

The “Compagnie d’Amenagement des Coteaux de Gascogne” (CACG) iscommissioned by the state to establish and operate the so-called “franchiseperimeters” for irrigation (“périmètres en concession”). These perimeters coversome 70,000 ha and constitute the main part of the irrigation areas managed byCACG. The franchise covers a 10-year mandate to operate and maintain theirrigation infrastructure. After this period, the water users can decide whether ornot they decide to extend the contract or vote for its closure.

The service to the water users consists of providing water at previously agreedflow rates and pressure to the field hydrants of individual water users. Thus, theCACG is responsible for operating all system components right through to theindividual point of withdrawal and also for any maintenance and repair work. Theseservices are secondary services to the primary service of water delivery to thefarmers and are summarised and indicated by the arrow S1 in figure M6-1.

These services are financed by the contractually agreed fees irrigating farmers paythe CACG. This return to the service provision is shown in figure M6-1 as thearrow f1. These fees cover the entire service package and are made up of a basicrate of French Franks 2000/l/sec and a per cubic-meter price of FF 30. The waterduty that has to be surrendered to the water agency constitutes a transitory itemand is forwarded by the CACG to the water agency. This flow of funds is indicatedby the arrow f3. Specific repair works may be contracted out by CACG to privatecompanies who provide these works as a service to CACG. These services arerepresented by the arrow S2 and the payment by f2.

The major coordination mechanisms that make these exchange relationshipsfunction are the following:

� A key coordination mechanism is to be seen in the franchise that the state hasgranted to the CACG. In Figure M6-1, this mechanism is marked as CM“concession”.

� A franchise agreement regulates the individual rights and duties of the CACGwhilst a “Conseil Administratif”, an administrative council ensures its correctinterpretation and implementation. This mechanism appears as “CM CA” in thefigure.

� Water provision itself is agreed with each individual farmer separately within theframework of a contract. This “contrat de fourniture d’eau d’irrigation” isindicated by CM1.

� The water users are members of formal or informal associations that arerespresented in the Neste Commission (CN). In this way they have thepossibility to discuss and negotiate general issues concerning the perimeterwith the CACG. This coordination mechanism is shown as CB ("collectivebargaining”).

� In the case of these perimeters, the function of “police d’eau” – enforcing rulesand applying sanctions with respect to water quotas – has been transferred bythe state to the CACG. CACG executes this role via state certified experts(“agents assermentés”). This mechanism is named CM “water police“ in FigureM6-1.

� The state pays certain premiums to farmers to encourage the use of watersaving equipment. The rules and preconditions for these premiums are laiddown in the “Politique Agricole Commune“. This is indicated by CM PAC.

� The relationships with contractors are governed by commercial contracts,represented by CM2 in the figure.

132

Module 6

The main three questions to be answered whenanalysing the governance of a service delivery systemare the following:

� Who provides which service(s) to whom?– Which services are being provided or are supposed

to be provided?– Who has the authority to arrange these services?

Fig. M6-1: Key services and service relations in the “périmètres

en concession”, i.e. irrigation franchise schemes in the Neste

System, Southern France (Source: Huppert and Hagen, 1999 in

MAINTAIN Case Study No. 2)

IRRIGATIONusers

STATEn Ministriesn DDE, DDAF, DIRENn Water Agency

CN

Companies

CACG

CM"concession" or

franchise

CM"CA"

CM"PAC"

CM"water police"

CM1

CM2CB CB

premiums

f3

f1

s1s2*

f2*

s = service provision

f = fees/finance

coordination mechanisms

Implementing a

Governance Analysis

of Service Provision

– Who is providing these services?– Who is receiving them and using/consuming them?

� What is being provided in return for each service?From whom and to whom is this return/paymentprovided?

� What are the coordination mechanisms and what arethe enforcing institutions that are making sure thatthese services and returns are actually being providedin a way that suits those concerned? In other words:what are the means available to service providers toinfluence the recipients or, conversely, the meansavailable to the recipients to influence the providerswhen it comes to upholding their respectivecommitments and obligations? What induces themnot to behave in an opportunistic manner?

Unless the answers to the first two questions can begiven without further analysis, they may be found withthe help of the above Modules 4 and 5, i.e. with theimplementation of the excercises of a “ServiceInteraction Analysis” (SIA) and of an “Analysis ofProperty Rights and Authority Systems”.

A “Governance Analysis of Service Provision” ingeneral and a “Governance Analysis of MaintenanceProvision” in particular may roughly proceed in thefollowing sequence:

Step 1: Elaborate a flow-diagram of the essentialservices

Based on the discussion of the “Hydro-InstitutionalService Chain” (see Module 4) the major actors areidentified that are involved in the provision problem athand. A flow-diagram of the essential exchanges isdeveloped,using the symbols described above.

Step 2: Incorporate the flows of the returns(finance, fees, compensations)

The respective payments or compensations for thedifferent services are discussed and incorporated into

133


the flow diagram (see Figure M6-1).An important pointhere is to be aware of whether or not a closed feed-backloop exists between the services provided – e.g.waterdelivery and maintenance – and the payments orcompensations for these services. Unless an externalfunding source is available that provides financialresources on a longterm and reliable basis, which inmost cases can no longer be assumed in irrigation,thesefeed-back loops between services and returns must beclosed if sustainable service provision is to be achieved.In fact, such a closed feed-back loop can be seen as acoordination mechanism in its own right: a direct linkbetween the service provision, and its quality, and thefees or prices paid for it. It is an important way to giveleverage to the client and to provide incentives to thosewho deliver the serves.

This is why institutional arrangements for irrigationfinancing are discussed in more depth in Module 9.

Step 3: Discuss and visualize the respectivecoordination mechanisms

Discuss with the relevant representatives of theprovider and the “client” the following questions:

� Who or what is making sure that the indicatedservices and returns are actually provided in a waythat is in line with the agreement between thoseparties? In other words: What kind of coordinationmechanisms are on hand for service delivery?

� To what extent can the service providers influencethe recipients or, conversely, the recipients theservice providers,when it comes to upholding theirrespective commitments and obligations?

The respective mechanism is then indicated with anumber or a letter in the “diamond”shape of the sketchthat is shown between the provider and the recipient.

Simple as this excercise may appear, the experienceof MAINTAIN has shown that the basic questions askedin the course of this step are seldom asked in practice.

134

Module 6

Hence, in many of the MAINTAIN Case Studies it wasfound that coordination mechanisms were eitherdeficient or did not exist at all.

Step 4: Analyse prevailing problems withcoordination mechanisms

Uncovering the problems related to coordinationmechanisms in a workshop setting – so goes the ex-perience gained in the course of MAINTAIN – normallyis an easy excercise if both the provider and the clientof a particular exchange relationship are involved in thediscussions.Both sides are the “experts”when it comesto analysing the other side’s failures and neglect.The questions to be asked are simple:

� Can the existing coordination mechanisms make surethat the provider sticks to the service agreement andprovides high quality service? What can the recipientdo, in case the provider tends to behaveopportunistically?

� Can the existing coordination mechanism ensurethat the recipient honours the service agreement andprovides timely and sufficient returns for thedelivered service? What can the provider do, in casethis is not so?

� Are the existing service relationship and thecoordination mechanism(s) such that they conveyincentives to the provider to deliver high qualityservices?

� Are the existing service relationship and thecoordination mechanism(s) such that they conveyincentives to the recipient to engage in the servicerelationship?

(to answer the last two questions, reference may bemade to Module 10). If there are no clear and positiveanswers to these questions, coordination mechanismsare too weak or non-existant. When remedies tocoordination problems are discussed, the followingsituation specific considerations deserve attention.

135


136

Module 6

Approaches to solve coordination problems will havea different focus, depending of the overall institutionalcontext. With reference to Chapter 4 and to Figures 5and 6, we describe the major thrust of such situationspecific approches as follows (see Figure M6-2).

Situation A:“Refining complete contracting”

In situation A, we deal with well-establishedinstitutional arrangements and with an institutionalframework that is supportive to these arrangements.Ideally, there should be no governance problems.However,in reality,steps 1 to 4 of the above analysis mayreveal governance deficiencies.

We refer here mainly to conditions in industrializedcountries. In such contexts, data availability, technicalknow how and institutional capacities are such that inmost cases detailed “asset management”approaches tomaintenance (see Module 2) will be feasible. Theseapproaches make available detailed quantitativeinformation concerning the technical requirements,thenecessary standards, the timing and the cost of theenviseaged maintenance program. Hence, on the basisof this information, it will be possible to specify clearlywhich services and returns have to be provided at whatpoint in time.These specifications can be clear enoughto be objectively verifiable ex post by third parties. Insuch cases we speak of “complete contracting”: thelevel and the quality of the maintenance service as wellas the price are unambiguously defined.4

Governance deficiencies in these situations willnormally refer to insufficient specification of thesesupposedly “complete” contracts. Solutions will hencetry to improve the terms of contract for themaintenance provision, possibly by improving thedatabase through better asset management approaches.This is why we speak in this case of “refining completecontracting”.

Toward situation

specific solutions to

governance problems

in maintenance

provision

4 For more details on types of contracting see supporting document MAINTAIN Thematic Paper No. 10

Situations B and D: “Coping with incompletecontracting”

In situations B and D, as they are defined in Chapter4 and Figure M6-2, institutional arrangements formaintenance provision are weak or non-functional.Normally, this will mean that asset managementapproaches are not a realistic option here. As aconsequence,the maintenance services that have to beprovided over a longer period of time cannot bedescribed in an exhaustive manner in advance. Also,since institutional arrangements are still evolving,maintenance objectives may change over time.In orderto allow the involved parties to react to such changesin a mutually beneficial way, contracts between themhave to be appropriately flexible.At the same time, thecontracts should not be open to opportunisticexploitation by either party.MAINTAIN Thematic PaperNo. 10 discusses a range of contractual provisions thatallow coping with such uncertain contract conditions.The major provisions of this sort are:

� A third party as arbiterThe service provider and the client can agree to usean arbiter in the event of a dispute. The decision ofthe arbiter has to be honoured by both parties.Otherwise,specified sanctions will be applied.Thesesanctions must also be part of the contract.Establishing or appointing an arbiter is particularlyuseful in environments where unambiguous laws arenot existing, or where neutral law enforcementcannot be taken for granted.

� Unilateral decision making authorityHere, the client side pays a premium to decide at agiven point in time whether or not the contract willbe executed or extended. For example, an irrigationorganization or a water user association may award alonger-term maintenance contract to a privatecompany (the longer term being the “premium”paid)but retain the unilateral right to cancel or to extendthe contract after a fixed time period.This provides

137


138

Module 6

incentives for the firm to deliver a good and reliableservice – even if not all the details of this service havebeen specified in advance – hoping that the contractmay be extended.

� StandardsGiven the difficulties in setting up a detailed assetmanagement plan, an irrigation organization or awater user association may decide to set a standardwith respect to the level of the water delivery service.Decisions on the details of maintenance provisionsare left to the contractor. As long as thepredetermined level of water delivery service isassured, the contract is fulfilled. Provenunderachievement will be penalized by certainreductions in the returns for the service in question.However, the setting of standards alone is anambiguous approach with respect to maintenance.The effects of neglected maintenance often appearonly in a distant future.Hence,the contractor may usethis fact to his/her advantage and reap the benefits ofthe contract while at the same time underinvestingin actual maintenance provision. This is why thisapproach normally should not be a “stand alone”solution.

Box M6-2: Using external arbiters in maintenance management

In Pakistani irrigation schemes farmers had been expressing concerns aboutthe quality and quantity of the maintenance that had been provided bycontractors on behalf of the Provincial Irrigation and Drainage Authority(PIDA). In 1997, the government decided to involve a few reputed firms tomonitor the maintenance work. These arbiters discovered that weaknesses inthe contracting procedures of PIDA were causing the maintenancedeficiencies. PIDA officers had specified the terms of contract fairly rigorously– trying to establish _complete‘ contractual arrangements – and had left littleflexibility for the contractors to adjust to unforeseen circumstances. Given thehigh negotiation costs for the contractors, the small quantity of work percontract, and the fear of loss of business, the contractors followedinstructions closely, even in situations where this proved to be technologicallyinadequate (from UL Hassan in MAINTAIN Country Paper No.2).

139


� Earmarking of maintenance expendituresIn situations B and D, it will often be difficult to planthe necessary annual maintenance activities inadvance. In context B, where the institutionalframework is supportive,the service provider can beobliged by external institutions (e.g.the government)to be prepared for unforeseen events and providesufficient maintenance.To achieve this,the state mayrequire that a “maintenance reserve” is earmarked inthe annual budget of the provider. This reserve isdesigned high enough to cover the costs of the worklikely to be necessary and cannot be used for otherpurposes. The state may stipulate an upper and alower limit each year for the maintenance of mediumand large-scale irrigation schemes.This will rule outany excessive preventive maintenance whilst alsoensuring that larger maintenance jobs are not put offfor years to come.The example of such a maintenance reserve ispresented and discussed in MAINTAIN Case StudyNo. 2, referring to water management systems inSouthern France.

� “Agreement to agree”In situations of insufficient information about futureevents and related maintenance and repair

Box M6-3: “Agreement to agree” arrangements in Andeanirrigation schemes

“Agreement to agree” contractual arrangements for maintenance andespecially for repair can be found in many farmer managed irrigatiuonsystems. This often escapes the notices of outsiders. In many Andeanirrigation systems, for example, there are implicit ‘agreement to agree’ ruleswith respect to unforeseen upcoming maintenance and repair needs.Irrigation farmers in the Bolivian Andes, for instance, adhere to the principleof ‘ad-hoc working groups’. To cater to changing and unpredictable repairrequirements, the irrigation farmers keep advance planning to a minimum butmaintain a common understanding (an ‘implicit contract’) to form ad-hocworking groups, whenever need arises.

requirements, a useful contractual provision may bethe so-called “agreements to agree”.In such cases,therespective parties arrive at an agreement – possiblydocumented in a contract – that requires to reach anagreement within a given period of time after theneeded information becomes available. Otherwise,specified sanctions will be applied.

� “Shot-gun clauses”In its simplest form,this is a coordination mechanismreminiscent of the distribution rule “one personslices the cake,but the other can choose which slicehe or she wants first”.Such an arrangement providesstrong incentives to the party responsible for settingup the distribution to strive for the most equitableway to do so, since otherwise the rule “backfires”against the decision-maker himself.

The examples of special contractual arrangementsgiven above illustrate how to cope with “incomplete”contractual rules. Using such mechanisms, the partiescan ensure that agreements are honoured and thatopportunistic behaviour is prevented, even in anuncertain institutional environment. With these andsimilar mechanisms it may be possible to establishgovernance for maintenance provision which isindependent of legal enforcement. Thus, suchmechanisms are likely to work even in countrieswithout a “functioning” (in a western sense) judiciary,as will be the case in situations of type C and D.

Situations of type C:“Strengthening solidarity”

Here we refer to type C situations, as described inChapter 4 and in Figure M6-2. In irrigation, these arefunctioning farmer managed irrigation systems. Verysmall systems of this kind will be subject to thegovernance mode of “collective decision making inprimary groups” as mentioned in Table M6-1 above. Insystems like these,there is strong social cohesion basedon local customs and traditions.Hence,trust,reciprocity

140

Module 6

141


and solidarity play leading roles as governancemechanisms.

The paradox we find here in relation to ourdiscussion on governance is the fact that many of thepreviously mentioned contractual mechanisms can becompletely counterproductive in this situation.Normally in this context the actors are farmers whohave supported each other for generations on the basisof customary “implicit contracts” of mutual help andreciprocity. A sudden switch to formal, and possiblyeven written, contracts may seriously erode mutualtrust and solidarity. Therefore, the challenge in suchsituations is to reinforce existing institutionalarrangements and traditional governance systems.

The important point here is, that the erosion of trustand solidarity mechanisms must be perceived as a veryhigh cost related to maintenance provision.This is whythere is ample evidence that many farmer managedsystems prefer to have some hydraulic structures builtsuch that they require frequent rebuilding or

Figure M6-2: Situation specific solutions to governing

maintenance provision

A B

C D

“Refining completecontracting”

“Strengthening solidarity”

“Coping with incom-plete contracting”

“Coping with incom-plete contracting”

strong

Service Arrangement

weak

Supportive

Unsupportive

External institutio

nalE

nvironm

ent

rehabilitation.Temporary stone weirs and groynes thatare washed away with peak runoff of the rivers and haveto be rebuilt annually are a case in point.The joint effortof the community for reconstruction contributes tostrengthen solidarity among the community members.The high costs of repeated investment of time andlabour are balanced by the benefits of communitysolidarity.

This comes down to the requirement to studycarefully the existing governance mechanisms insituations of type C, so as to devise approaches tomaintenance that strengthen solidarity mechanismsinstead of eroding them.

142

Module 6

Module 7

Analysis of “Principal-Agent” problems(Supporting Documents: MAINTAIN Thematic PaperNo.10 and MAINTAIN Case Study No.3).

Principle-Agent problems are deficiencies related tocontracts and agreements between exchange partners,in our case between the provider and the client of aservice relationship.In another terminology,as we usedit before, such problems are particular deficiencies of“coordination mechanisms”. Problems of this kind aredue to the fact that the provider side has moreinformation about the service provision process thandoes the client side. Such a so-called “informationasymmetry” is, on the one hand, both necessary anddesirable since it reflects the division of labour and thespecialisation of the provider. On the other hand,though, the actor who is not as well informed, the“principal” (in our case the client of the servicedelivery) runs the risk of being exploited by the betterinformed party, the “agent”– the service provider.

Maintenance is a typical “future good” (see Chapter2.3 in Part One of this Guide).This means to say that thebenefits of maintenance normally accrue in a moredistant future.When,for example,the irrigation agencyor farmers do regular preventive maintenance at acheck-structure and intake sluice, the benefit is thatthey avoid deterioration or possible failure of thesestructures some years from now.This characteristic ofmaintenance makes it difficult for those whocommision maintenance services, the principals, toreceive immediate proof that the service has been doneaccording to agreement unless they invest heavily inmonitoring efforts. On the other hand, those whoprovide the maintenance service, the agents, may usethis “information asymmetry”with respect to the actualwork done,to behave in an opportunistic way.They may

143

Analysis of “Principal-Agent”problems

What are “Principle-

Agent” problems?

Why is the analysis of

Principle-Agent

problems important in

the context of

maintenance service

provision?

be tempted to provide suboptimal service (or even noservice at all) without running a big risk of being heldaccountable. This is why it is important to look moreclosely at the types of Principle-Agent problems thatmay be prevalent in a given situation and devise con-tractual measures that can counteract such deficiencies.

The major Principle-Agent problems to keep an eyeon in the context of maintenance service provision arethe following:

� The “Moral Hazard”problem� The “Hold-Up”problem� The problem of “Adverse Selection”

The “Moral Hazard”problemA moral hazard risk arises whenever two actors are

joined in a client-supplier relationship. The client(principal) commissions the supplier (agent) toperform a service on his behalf and thus confers acertain scope for decision-making on the supplier.If wepresume that the agent’s activities cannot be directlymonitored by the client, and that the agent makescertain observations during the execution of the orderwhich the principal has not made, then this leads to anasymmetrical information status between the twoactors concerned. If it is also presumed that the task isso complex that it can be influenced by many externalfactors, the following problem can arise.Following thesigning of the contract, the agent might reduce hisefforts to fulfil the terms of reference (reduce his cost),without the principal being able to call him to account.The agent can always claim that a poor result is due tocircumstances beyond his control,thus relieving him ofany guilt or responsibility.

144

Module 7

What are different

types of Principle-

Agent problems that

may arise in

maintenance service

provision?

The purpose of this Module is to describe different types of potentialPrinciple-Agent problems, to discuss the consequences they have onmaintenance service provision, and to introduce some ways to cope withsuch problems.

145


The “Hold-Up”problemHold-up problems arise in service relationships in

which the potential service recipient has engaged insignificant prior investment in the service to beexpected. In such situations, the potential recipient isheavily dependent upon service delivery and on theservice provider. In case the client (the principal) doesnot receive the expected service, all his investmentmight be in vain.The danger exists that such a unilateraldependency may be exploited by the service providerin order to extract particular favours (“sidepayments”)from the client. The client may also take advantage ofthe dependency by negotiating a premium fee forservice or by cutting costs and providing sub-optimalservice. This danger is particularly high if stronginformation asymmetries exist between supplier and

Box M7-1: Facing Moral Hazard problems in maintenanceprovision – an example

A water user association or an irrigation district has established aprofessional maintenance unit. The staff of that unit may be asked by theboard of the association to prepare a asset management plan (see Module 2in this Guide). In doing so, the staff might, on the basis of its professionaljudgement, make the interim decision to replace certain infrastructure partswith new equipment. Additionally, they may also decide to plan a level ofmaintenance well above the “real” requirements. In doing so they behave –conciously or unconsiously – in an opportunistic way, trying to secure theirown positions and maximize the funding for their unit. In presenting itsmaintenance budget requirements to the board, the professionals takeadvantage of the information asymmetry faced by the board. Given all thetechnical details and judgements involved in developing the overall plan,information that cannot be available to the decision makers, board membersmay be unable to understand or change underlying decision criteria. Thesituation is analogous to that of a patient of a highly specialised doctor whocan hardly judge the therapy if he does not even understand the diagnosis.

The board faces a Moral Hazard problem: Shall it trust the professionalcompetence of its maintenance unit or will it be better to assume (at leastsome) opportunistic behavior of this unit and therefore reduce the fundrequest when deciding upon the budget to be allocated?

146

Module 7

recipient and if external conditions beyond theinfluence of the provider may cause high uncertaintyabout the level of service that will be provided. Theprovider can then blame these conditions as an excusein case of non-delivery or sub-optimal provision of aservice.

Box M7-2: Facing a Hold-Up problem in maintenance provision –an example

A typical Hold-Up problem in irrigation arises where the supply of equipmentis implicitly (but not formally) coupled with the subsequent provision ofmaintenance for this equipment. The supplier of a particular set of pumpinggear, for example will, in most cases, be asked to perform maintenanceservices at least for specific maintenance tasks that require specialist know-how that only the supplying company possesses. In cases like these, theprincipal, the irrigation agency or the water user association, might have toface a Hold-Up problem. Since the principal has made a specific investment,buying a special pumping set that is only available from a particular firm, thisfirm may use the situation to its advantage and charge excess prices formaintenance service.

The problem of “Adverse Selection”An Adverse Selection Risk exists whenever two

actors are planning to join in a client-supplierrelationship. The client (principal) intends to buy aservice or a good from the supplier (agent). Theremight, however, be characteristics of the service orgood which are unobservable to the client but knownto the supplier.The supplier might not have an interestin revealing any information about thesecharacteristics.As a result,the buyer might find himselftrapped in an unfavourable exchange relation aftersigning a contract, or he might not enter therelationship at all because he anticipates being takenadvantage of. Both of these outcomes can result inmissed benefits for all partners. Better selectionmechanisms (signalling and screening) offer solutionsto these problems.

There are three aspects that are of importance here:

� The first one is that problems of “Moral Hazard”andof “Hold-Up” are most relevant in relation tomaintenance problems.

� The second one is that we cannot strictly separatewater delivery and maintenance when we try toaddress these problems. This is because Principle-Agent problems in water delivery may well be theroot cause for deficiencies in maintenance.

� The third aspect is that we need to realize thattransparency and accountability are key terms inthis context.

Considering these three aspects, we may use thefollowing steps in order to detect Principle-Agentproblems in maintenance service provision.

Attempts to improve service relationships by solvingPrinciple-Agent problems generally take one of twodirections. They either try to minimize the existinginformation asymmetry or they strive to bring about acoincidence of interests between the principal andagent through changes in the incentive systems.In mostcases attempts to minimise information asymmetrieswill entail high monitoring costs. In many situationssuch attempts may upset existing divisions of labour

147


How can we detect

Principle-Agent

problems in irrigation

maintenance?

Box M7-3: Facing problems of “Adverse Selection” – an example

As stated above, adverse selection occurs when, before a contract has beensigned. The buyer faces several options but runs the risk to make asuboptimum selection due to problems of intransparent information. Oneexample from irrigation maintenance relates to the reluctance of internationaldonors to fund long-term maintenance costs. Maybe one of the reasons isthat they are aware of the difficulties involved in monitoring maintenanceefforts. Thus, the donors refuse to commit themselves to maintenancefunding in most cases, often incurring substantially higher rehabilitation costsinstead.

What are solution

approaches for

Principle-Agent

problems in

maintenance service

provision?

148

Module 7

and responsibilities and hence create greaterinefficiencies. Changing incentive systems such thatthey bring about coincidence of interests may alsoresult in substantial costs. The basic idea of creatingincentives in service delivery systems is to rewardproviders in direct relation to their contribution to theservice provision process. It is then in their personalinterest to make improvements in service provision.This is why a full or partial financial autonomy for theprovider seems to be of prime importance. Thequestion then is to find institutional arrangements that

Detecting Principle-Agent problems in irrigation maintenance

1. Analyse the service delivery system for water and identify deficiencies inprimary or supporting services (Follow Module 4 in this MAINTAIN Guide).

2. Taking these cases of deficient provision of services or supportingservices, find out what are the contractual mechanisms that govern thatrelationship (Follow Module 6 in this Guide).

3. Discuss with the involved parties whether or not these mechanisms ofcontractual governance create transparency for the principle or whethersubstantial asymmetries of information exist.

4. Discuss with the involved parties whether or not there are unpredictableexternal influences that may effect the quality of the service provision, andto what extent this has impacts on the accountability of the provider(agent).

5. In cases where both a lack of transparency and a lack of accountabilityare found, a problem of “Moral Hazard” will be most likely and thesituation might invite the service providers to behave in an opportunisticway. They may face the temptation to use the situation to their ownadvantage, e.g. for rent-seeking purposes (With respect to Rentseeking inthe context of maintenance service provision see MAINTAIN ThematicPaper No. 9).

6. Organize workshops or round-table discussions to find solutions to suchproblems, using the orientations given below.

7. In case no such problems can be detected in the context of water delivery,proceed to maintenance service provision and follow the sequence ofsteps 1 to 6 as mentioned before.

149


bring about motivation for the provider to “search forexcellence” in service provision.

Orientations for possible solutions with respect toPrinciple-Agent-Problems are summarised in Table M7-1 and discussed more in detail in the following text.

Table M7-1 : Types of Principle-Agent problems and solutions in servicerelationships (see MAINTAIN Thematic Paper No. 10 and MAINTAIN Case Study No. 3)

Adverse Selection Moral Hazard Hold up

Type of Risk of a suboptimal Risk of suboptimal Particularly strong risk of “Moral Problem selection of a service service provision due Hazard” due to pre-service

provider/agent by the to opportunistic investments incurred by theclient /principal behaviour of the client / principal and resulting

provider / agent, who, crucial dependence of thehowever, cannot be client from the service held accountable provision

Origin of Information Information Information asymmetryProblem asymmetry asymmetry coupled with

one-sided dependency causedby specific pre-service investments

Causes Qualification of service Detailed activities of Detailed activities of provider/ behind provider / agent and provider/ agent and agent and external influences Infor- quality of service external influences on these activities not knownmation provision not known on these activitiesAsymmetry not known

Time when ex ante ex post ex postProblemis Acute

Theoretical Create/improve Create/improve Vertical integration or creation of Approaches selection mechanism incentive systems mutual dependenciesto Problem that counteract Solution “Moral Hazard”

Examples � “Signaling” or � Improve return/ � Create joint property of exposing proof of compensation for resources for agent and qualifications or service delivery principalinformation on � Team building � “Exchange of hostages” orservice delivery � Manipulation of handing over security

“Outside Options” to principal

According to theory the major paths for solutions toPrincipal-Agent problems in service relationships arethe following:

1. “Moral Hazard”: (Again: This is the risk ofsuboptimal service provision due to opportunisticbehaviour by the provider, who however cannot beheld accountable by the client.)

Approaches to solutions may be the following:� “Performance Compensation”

In this case the client/principal tries to couple thelevel of returns/compensations for the service toverifiable indicators or other facts that can beinfluenced by the provider/agent. One way may beto agree upon individual performance levels, as inManagement by Objectives.

� “Tailoring” the scope of the service to be providedIn cases where a provider/agent is supposed toprovide several services that make use of the samepool of resources, she may be tempted to findexcuses for a sub-optimal level of service provisionby pointing to problems with the provision ofanother service. She also may have an interest indiverting resources from the provision of one serviceto that of another. This can be counteracted by“tailoring”well-defined service orders and allocatingcorresponding budgets to them.

� “Team-Building”Information asymmetry may be avoided throughteam-building efforts where team members eachhave an interest in preventing “moral hazard”actionsby others.Social pressure and control can thus bringabout an efficient provision of service.

� “Decentralisation of Ownership”The idea here is, that in cases where a local actor isthe owner of certain resources, she may treat theseresources with more care than she would dootherwise. Hence, one may try to transfer certaincritical resources to the ownership of this actor.

150

Module 7

� “Efficiency Pricing”In cases where a client (principal) can get hold ofservices on the basis of agreements for returns(payments) that do not reflect the scarcity of theservice, she will tend to overuse and waste theservice. Hence “efficiency pricing” is needed toprevent such Moral Hazard problems.

� Manipulation of “Outside Options”.In this case, the client/principal tries to stopundesirable “side-actions”by the provider/agent thatmay have an effect on the service in question.

In the case of the Moral Hazard example mentioned inBox M7-1,the following approaches (and combinationsthereof) might be considered.� One or two of the board members may join the

maintenance planning process.This corresponds tothe “team building” approach, and also tends toreduce information asymetries.

� The board may opt for a better “tailoring” of themaintenance service (see above) and separateresponsibilities, for example, for irrigationmaintenance and for the maintenance of vehiclesand equipment which were combined. This mayrestrict possibilities for the maintenance staff todivert resources from one service to the other andfind excuses for sub optimal levels of both.

2. The “Hold-Up” Problem: (Again:This is whenthere is a particularly high risk of “moral hazard”due to highly specific pre-service investmentsincurred by the client and resulting dependence bythe client on the service provider.)

Approaches to solutions may be the following:� “Joint ownership of resources”

The dependence of the client/principal upon theagent may be avoided through the creation of jointownership of resources.Thus, both partners will bemutually dependant on a functioning serviceprovision.

151


� “Exchange of hostages”The provider/agent may hand over some kind ofsecurity to the client/principal to convince her of hisgood intentions and to transform the unilateraldependency into a mutual one.

� Avoidance of high specificity of investmentIn this case, the “principle” will try to specify herinvestment needs in such a way that there may beseveral potential offers at the market.

In the case of the irrigation maintenance example fora Hold-Up situation,mentioned above in Box M7-2, thefollowing approaches (and combinations thereof)might be considered.The irrigation agency or wateruser association (the principal) might opt for morestandard types of pumping gear in the future for whichthere are multiple offers on the market (Avoidance ofhigh specificity of investment)� When buying the pumping equipment, the

principal might insist on some kind of servicecontract that is coupled to the buying contract. Inthis case, a future Hold-Up situation might beavoided (Exchange of Hostages).

3. “Adverse Selection”: (Again:This is the risk ofsub-optimal selection of a service provider by theclient due to information asymmetries.)

Approaches to solutions may be the following:� “Signaling”Activities

To avoid “adverse selection” of a service provider(agent) by the client (principal),the provider may tryto engage in “signaling” activities. She can try toprovide proof of her ability to provide a service ormay try to make the quality of her service provisiontransparent to the client

� “Screening Mechanisms”While the costs of “signaling”activities are borne bythe agent, the principal may proceed to collectinformation on the qualifications or service provision

152

Module 7

capacities of the agent. She may do this using so-called “Screening-Mechanisms”. She may look forreferences from third persons, carry out tests, orengage in other information collection activities.

In the case of the irrigation maintenance example foran Adverse Selection situation described in Box M7-3,the following approaches (and combinations thereof)might be considered.

Donors might collect better data on existingmaintenance practices in particular countries. Theymight then make future agreements to the provisionof funds for new irrigation projects or forrehabilitation measures – including some initialfunding for maintenance during a “start-up phase”–subject to favorable maintenance ratings.(“Screeningmechanisms”).

Recipient countries might engage in similarexercises and “signal” their (hopefully) positivemaintenance strategies and records.

153


Module 8

Free Riding – A central governance problemrelated to maintenance(Supporting Documents:MAINTAIN Case Studies No.3and 7, MAINTAIN Thematic Paper No. 9; GTZ-publication No.263).

As mentioned in chapter 2.3 of Part One of thisGuide, maintenance in many cases is a so-called“collective good,”or more precisely, a “club good.”Thismeans that the maintenance service in most instancesis not provided for an individual but for a group.Whereit is not possible to exclude non-payers (or “free riders”)from receiving the same maintenance services aspayers,we say that there is a problem of “open access”or “non-excludability” of the service. By contrast,effective governance by an irrigators’ group willinstitute arrangements that prevent free riding.

Free riding leads to a serious “feed back deficiency”between the service provider and the client. Theprovider cannot exclude non-paying demanders,and inturn the individual clients, the water users, loose theirpayment power to enforce high quality serviceprovision.This normally entails a serious breakdown ofincentives to further service provision:even those whoare willing to pay for the service will discontinue to doso when they become aware that free-riding isincreasing. With increasing free-riding the financial

154

Module 8

What is “free riding”

and what are causes

leading to it?

The purpose of this Module is to explain the causes for problems of freeriding and to highlight its importance as a prime “motivation killer” ordisincentive for the provision of maintenance services. The Module points tosome options for the solution of free riding problems.

155

Free Riding – A Basic Governance Problem related to Maintenance

Box M8-1: Free-riding behaviour in maintenance provision – arationality trap

Where maintenance is provided by a community of water users as a clubgood, the individual water user is faced with a conflict between his owninterests and those of the community. This conflict – social scientists speakof a “prisoner’s dilemma” – can damage the community interests, and thusultimately also the water user’s own interests.

The fact that a water user who fails to make his contribution cannot beexcluded from the benefits of maintenance tempts individuals to think alongthe following lines:

The economically most beneficial situation for the individual arises when allother individuals contribute towards maintenance, but the individualconcerned plays the “free-rider” by profiting from the maintenance withoutmaking any contribution. The profit he derives from irrigation is thus – unlikefor the other users – not diminished by the costs of maintenance inputs fromhis side. The individual sees the damage which he inflicts on the communityby withholding his contribution as minimal, in view of the large number ofcontributors.

The economically most disadvantageous situation for the water user inquestion arises when he pays his contributions in full, whilst everyone elsetries to act as a free-rider. The individual’s profit is then reduced on the onehand by the fact that no maintenance or only sub-optimal maintenance canbe carried out on the system. Secondly, the water user’s profit is reduced bythe fact that he, unlike everyone else, has contributed towards maintenance.

The economic conditions which emerge when all users, including theindividual considered here, make their contributions, are calculated by theindividual as falling between the two aforementioned extremes. The addedprofit is confined to the additional benefit resulting from the now fully-maintained system, minus the contributions paid by the individual. The wateruser will accept this “second-best” solution unconditionally if and when hecan be certain that everyone else will act in the same way. However, once thisbecomes uncertain, the picture changes:

In a situation like this where it cannot be ruled out that free-riders will go“unpunished” for profiting from community efforts (which is often the casewith larger systems and where there is a lack of legal security or wheremeans of imposing sanctions are inadequate) there will therefore be atendency for individual users to withhold their contributions “just to be on thesafe side”. Users will do this especially once the contributions becomedisproportionately high in relation to the overall benefit. This, however, willundermine the morale of those actually making their contributions.

156

Module 8

The number of contributions made will fall, until the burden of payment onthe remaining cooperating users becomes so great that they too stop paying.The damage inflicted on all by non-maintenance of the system now falls backon the individual. His calculating attitude turns out to be a rationality trap, inthat the second-worst situation arises: Whilst the individual has saved hisown contribution, the (additional) benefit from maintenance has notmaterialised. Instead, significant damage has been suffered by all. Therationality which had been based on the individual’s own interest has turnedout to be a trap.

The reader ought to be familiar with such prisoner’s dilemma-type situationsfrom his or her own experience, even though he or she may not have anyexperience of being a water user. The car driver who knows thats he shouldstop driving in order to avoid major detriment to the community in the form ofair pollution is in a comparable situation. Calculating her own cost-benefitratio, the driver will assume that the benefit which would be derived by thecommunity if a single car – i.e. her car – were to be driven less, would beminimal. In other words, it would only make sense to stop driving if he couldbe certain that most other drivers would act likewise. However, since thelevel of uncertainty in this case is high, she will attempt to avoid the situationmost disadvantageous to herself – namely that everyone else carries ondriving, but not her, and the environment continues to be damaged. Sincemost other drivers will also do the same thing, damage to the environmentcontinues – up to a point at which the community and therefore the individualsuffer severe damage.

This example, which is not related to irrigation, is meant to demonstrate thatprisoner’s dilemma situations can occur as problems in the control ofcollective goods in a wide variety of social systems in which large numbers ofindividuals are engaged in community activities.5

5 A basic discourse on this theme was first produced by Mancur OLSON,bearing the revealing title of “The logic of collective action” Cf. OLSON,1968.

basis for the provision of the service will erode andprovision will eventually stop.

Strategies that intend to overcome free ridingproblems related to maintenance will vary with thekind of institutional arrangements that govern therelationship between the provider and the client in theservice relationship (see Module 6 of this Guide):

How can free riding

problems be

overcome?

Scenario 1: Maintenance as a pure market service

This is a situation where maintenance is provided ona contract basis “on the market” to a private farmer orto the government. The maintenance service in thesecases is a so-called “private good”. Potential problemsrelating to collective goods, and the associatedrationality traps,do not arise in the service relationship.

Scenario 2: Maintenance as a provision by and fora collective entity

Where maintenance is provided by and/or to a groupof farmers, free riding may occur. It is appropriate todistinguish between two basic types of collectivesystems. These are the relatively small primary groupand the larger self-administered systems (which aretoo large to enable all actors to interact directly on aface-to-face basis) (see Module 6). Solutions to freeriding are likely to differ in the two contexts.

With primary groups the strategy to overcometendencies towards opportunistic behaviour arisesmore or less of its own accord. All members of thegroup know each other and have close socialrelationships with each other. As a result, the socialpressure on the individual member is so great that free-riding almost never occurs,or if it does then it leads toserious social sanctions. From the point of view of theindividual member, rationality traps are virtually non-existent in primary groups.

By contrast, in larger self-administered systems theproblem of rationality traps does arise. In this case,social sanctions are very difficult to apply. It is no longerthe case that each member is kept informed as to theactivities of each of the other members. Shielded by theanonymity of the large group, it is usually possible toindulge in opportunistic behaviour without anydifficulty.

The following strategies may help to counteract freeriding behaviour in this context.

157


The “small group strategy”The “small group strategy” involves keeping the

group which administers itself as small as possible,restricting it to the size of a primary group. To put itanother way, application of the small group strategyamounts to the formation of smaller sub-groups whichare just about capable of functioning on the basis ofmutual social responsibility. Important factors here arethe homogeneity and cohesiveness of the group. Theweaker they are,the smaller the group size needs to be.

“Moral persuasion”A further strategy in collective systems involves

“moral persuasion”. The aim here is to ensure that free-riding is stigmatised within the community as anti-social, becoming the subject of moral disapproval.However, there are strong constraints on moralpersuasion, especially in larger groups, and where theshare of opportunity costs for positive social behaviouris high.

“Selective incentives”Another strategy under the given institutional

conditions here involves “selective incentives”.Selective incentives are goods or personal advantagesdesigned to induce potential free-riders to make theircontributions to the collective good, e.g. tomaintenance.To achieve this they are offered additionalbenefits, the receipt of which is dependent on theircontribution to the maintenance effort. Anyone whofree-rides under these conditions can be excluded fromobtaining the additional benefit. One problem withselective incentives consists of the fact that the benefitin question needs to be financed from the contributionsto the collective goods. Consequently, they are tied toa certain level of contributions. In addition,the benefitmust be offered on better terms (such as more cheaply)than on the market. Neither of these conditions withrespect to selective incentives is easy to meet in water

158

Module 8

user organizations (e.g., application of special materialallocations or special services).

”Coercion”A final strategy pursued for mitigating rationality

traps in collective systems is that of coercion. It isimportant here that this coercion is not applied againstthe actors’ will, but rather with their consent, in orderto provide everyone with the security that nobody willindulge in free-riding. The problem with this strategyis that in many cases this premise is in reality notfulfilled, or that some individuals impose their will onothers.

The coercion strategy may consist of cuttingirrigation water supplies in cases where maintenancecontributions have not been paid. In the “Valle Alto”irrigation scheme in Bolivia such a strategy is agreedupon by all participating communities who have to paythe whole share of their members’ fees in advance ofthe irrigation period, unless the supply to theircommunity fields will be cut.

Scenario 3: Maintenance within the scope of apublic utility

An irrigation organization which functions as apublic utility, and whose survival is dependent on ascomplete a collection of service charges as possible,willhave to link its maintenance provision to the service ofwater supply.Ideally it needs to create the technical andorganizational preconditions for water to be allocated,measured and charged to individuals or small groups.To provide water to larger groups that cannot solvetheir free riding problems will jeopardize the provider’ssurvival.

Scenario 4: Maintenance in state-administeredirrigation systems (central administration)

As a rule,this will involve large systems. Applying the“small groups” strategy here would mean forming

159


smaller sub-groups from the outset to function on aprimary group basis. For existing systems this will notalways be easy to achieve.

One important strategy here will be selectiveincentives, which the state organization can makeavailable to the water users who pay the water charges.

In reality, the predominant strategy with stateorganizations will be coercion. By means of directivesand corresponding monitoring, water users will berequired to practice “production under closesupervision”. Most large colonial and post-colonialirrigation systems have relied on this strategy.However,the information and monitoring activities needed toidentify and sanction free-riders as a rule involve highcosts.These costs are added on to the already high costsof state responsibility for operation and maintenance ofthe systems.It therefore comes as no surprise that manygovernments consider it appropriate to transfer somepublic systems to the water users. In this context it isoften overlooked that the size of the systems oftenmakes the strategies to deal with the problemsassociated with collective goods fairly difficult, evenwith self-administered systems.

160

Module 8

Module 9

Institutional Arrangements for IrrigationFinancing(Supporting Documents: MAINTAIN Thematic PapersNo.5 and 12;MAINTAIN Case Study No.7)

In every exchange relationship, where a good or aservice is delivered by a service provider there needsto be a “return” or a “payment” to compensate for theprovision. There are practically no such relationshipsthat can function without an appropriate or agreedreturn. Relationships in charity may be an exception,when altruism induces a person or an organization toprovide some goods or services and not ask forcompensation.Even in this case there is a whole strandof literature that discusses arguments that there are“invisible”returns like a good conscience,avoidance ofsocial sanctions, gain or maintenance of politicalinfluence,etc.

In the following we use the term “payment” for allkinds of returns, even if these returns have a non-monetary form.

The payment in an exchange relationship basicallyhas four functions:

� The payment is supposed to cover the costs of theprovision;

� Beyond cost-recovery, the kind and level of paymentconstitute the material incentive (or disincentive) forthe provider to engage in the exchange.They enablesubsequent exchanges between the partners tocontinue and hence are the bases for a sustainableexchange relationship;

� The payment creates accountability. The providerwill feel accountable to those who pay for theprovision. A closed loop between provision andpayment is therefore a key aspect of functioningservice provision.

161

Module 9

What is the

importance of

institutional

arrangements for

irrigation financing?

� The payment helps to create “customer sovereignty”.To have the capacity and the authority to pay – andto delay, withhold, reduce or increase payments – isan essential factor that gives both incentives andpower to the customer.

Institutional arrangements for financing would hardlybe a topic worth discussion if all exchange relationshipswould have the form of the bilateral closed exchangethat is prevalent in a simple commercial relationship.The service provider provides the good or service andthe customer simultaneously assumes the roles of“arranger,”payer and consumer of the service (see Fig.M9-1 on page 170). The arranging function hereincludes the function to select the provider, to decideupon the terms of the provision, to conclude anagreement or contract with the provider and to monitorthe provision.

Unfortunately, many exchange relationships,especially in public service provision, do notcorrespond to such a simple arrangement. Thefunctions of arranging,paying and consuming are oftenassumed by several actors. For example, the customermay be a needy individual requiring welfare supportand is only the consumer of the welfare service. Otherentities perform the functions of payer and arranger ofthe service (see Fig. M9-2 to M9-5 on pages 170 and171). Service provision now takes place in the contextof a “multi-actor” arrangement. The question arises,What are the rights, rules, regulations, agreements andprocedures (i.e.,the institutional arrangements) thatgovern such a service network of actors? Although wehave touched upon this issue several times before, theimportance of these institutional arrangements cannotbe overstated. Without well designed arrangements ofthis kind, service provision will not take place in aneffective and efficient way. This is a lesson which hasbeen learned by hard experience in irrigation serviceprovision in the past.

162

Module 9

The classical response to problems of maintenancein irrigation, a response especially given bydevelopment banks, has been to raise water fees.Withincreased returns from irrigation service provision (sogoes the argument) the major problem related tomaintenance (i.e., the lack of funds) can be overcome.

Implicitly, such a line of argument relates to a modelof service provision as it is shown in Figure M9-1 onpage 170. There is a closed feedback loop betweenservice provision and payment and hence shortages offunds will be eliminated by increased payments.

This Module – and much of the discussion of thisGuide – centers around the argument,that in most casesinstitutional arrangements for irrigation serviceprovision are complex. It is not only the insufficientlevel of funds but also the deficiencies in institutionalarrangements that are the root causes for maintenanceproblems.In many cases actual levels of water fees paidare strongly influenced by the institutional arrangementthat is in place.

We propose that the following key-questions beasked when trying to assess the viability andeffectiveness of incentives for a given institutionalarrangement for financing irrigation services:� Is there a closed feedback loop between services

provided and payments made? In other words, isthere a direct link between services and payments?

� Is full “customer sovereignty” assured? If not, whatrestrictions are placed on the functions of arrangingand paying for services?

163

Institutional Arrangements for Irrigation Financing

What is the purpose

of this Module?

The purpose of this Module is therefore to present a simple way for how toproceed when trying to assess the viability of institutional arrangements forthe financing of irrigation services, and hence for maintenance. This Moduleis applicable in all institutional contexts defined in chapter 4. However, it willbe particularly useful in context of type B, indicated in Figures 5 and 6.

The key questions to

be asked

164

Module 9

� Are subsidies provided? If yes,are they justified on thebasis of well-specified conditions related to servicequality?

In the following, we present and discuss five basicinstitutional arrangements for irrigation serviceprovision and financing which we consider to be themost frequently encountered situations. Doing so, werefer to figures M9-1 to M9-5. Our brief discussion ofthese cases centers around the above-mentioned keyquestions, from which we draw conclusions about theresults of incentives for both the service provider andcustomer. These cases are meant to guide practicalattempts of professionals to assess institutionalarrangements for irrigation financing. (Our rating ofincentive effectiveness goes from “+++”for “very high,”through “++” and “+” ratings for “high” and “low”respectively, to “-”as the rating for “non-existent”).6

Institutional Arrangement 1: Private ServiceProvision (see Fig. M9-1)

As mentioned before, this is the ideal case of anexchange relationship: the loop between serviceprovision and payment is closed, the customer has full“sovereignty”(i.e.,assumes the arranging,payment andconsumer functions) and no subsidies from a third partyare involved.

Such cases do not only correspond to commercialservice provision where the bilateral relationship isgoverned by market mechanisms.They also reflect thecase of self-governed community irrigation systemswhere the community organization provides irrigationservices to individual farmers who pay for such services(in monetary or non-monetary terms). Also, by way oftheir community membership and the mechanisms ofparticipatory decision making, farmers influence thekind of services to be provided and thus retain thearranging function. (The larger the group and the less

What are the basic

institutional

arrangements for

irrigation financing

and how can we

assess their potential

“incentive

effectiveness?”

6 For a more in-depht discussion on the preconditions for the effectivenessof incentives see Module 10 “Actor Specific Incentive Profiles”.

165


direct influence the users have in arranging the service,the less sovereignty they will preserve).

Incentive effectiveness:For the provider side : +++For the user side: +++

Incentive effectiveness can be ideal. In cases wherethe payment is appropriate and where Principle-Agent-Problems can be excluded (see MAINTAIN-ModuleNo. 7),the provider faces optimal incentives to providegood services.The provider is also fully accountable tothe users who retain full customer sovereignty.In casesof good service provision, the users will also have highincentives to maintain the service relationship.

Institutional Arrangement 2: Government Subsidyfor Private Service Provision (see Figure M9-2)

Cases of government subsidy for irrigation serviceprovision abound in developing countries. The casereferred to here and illustrated in Figure M9-2,howeveris special insofar as subsidies go directly to the userswho thus retain their customer sovereignty in therelation with service providers. This may be a case

Box M9-1: Government subsidy to private service provision – thecase of Southern France

Cases like these are found for example in the Gascogne-Region in SouthernFrance (see MAINTAIN Case Study No.2). Water user associations in thisregion – the so-called “associations syndicales authorisées” (ASA) – areentitled to various subsidies for irrigated agriculture from the state. Theyreceive large subsidies for the construction of irrigation systems and duringthe operation phase they benefit – amongst others – from subsidies for watersaving system components. However, after the construction and after aninitial one-year operation phase run by the state owned “Compagnied’Aménagement des Coteaux de Gascogne” (CACG), the ASA are free tochoose whatever service providers for operation and maintenance provisionthey prefer. The ASA enter into market based contracts with these providersand thus assume full customer sovereignty in the service provision process.

where water users receive funds from the governmentthat are completely free to enter into servicearrangements with particular service providers towhom they pay competitive prices.

In such cases, incentive effectiveness might be ratedas follows:

For the provider side : +++For the user side: +++

There is a direct and closed loop between serviceprovision and payment. The payment, arranging andconsumer functions are all assumed by the water usersand the subsidies are linked to clearly pre-determinedconditions that are linked to service quality (e.g.,procurement of water saving equipment).

Institutional Arrangement 3: Government subsidyand arranging function jointly assumed betweengovernment and users (see Fig. M9-3).

Also in this case, the government provides a subsidyto maintenance.However this subsidy is tied to certainpreconditions. Moreover, work quality may be subjectto government monitoring.

Such an institutional arrangement maintains a closedloop of service provision and payment between theservice providers – e.g. local artisans – and the users ascustomers. However, customer sovereignty here isrestricted. Being the paymaster gives the WUA somecontrol over maintenance service quality,however thearranging function is exercised jointly with thegovernment since the responsible government entitymonitors work quality in making payments to the WUAunder their contract. And finally, in the case discussedabove,government subsidies are not clearly earmarked.As mentioned before, one part is provided to eachnewly formed WUA,the other part is allocated on a per-acre basis.The subsidies have no link to service quality.

Conditions like these may justify an incentiveeffectiveness rating as follows:

166

Module 9

167


For the provider side: ++For the user side: +

The providers of the maintenance are directly paidby the WUA and nominally they are fully accountableto them.Hence,they may have high incentives for goodperformance.In reality however,they may be aware thatthe WUA do not have the full engineering competenceto judge work quality in detail.Given the partial role ofthe responsible government entity to monitor andapprove work quality, there may be cases wherecontractors can use this splitting of roles to theiradvantage.

The users have the incentive to take initiatives withrespect to maintenance, even if their sovereignty isrestricted by the government. However, the fact thatsubsidies are given without a clear earmarking toservice quality might induce users to go for “gold platedmaintenance,”since they perceive opportunity costs of

Box M9-2: Government subsidy and jointly assumed arrangingfunction

Examples for such cases are reported for the post-reform maintenanceservice provision in irrigation schemes in Adhra Pradesh, India (seeMAINTAIN Case Study No. 5). At the level of minor canals and below, wateruser associations (WUA’s) receive maintenance grants from the governmentto organize and carry out the work. There are two types of grants. The first isa block grant to each newly formed WUA for emergency repairs andmaintenance, to be used as the WUA sees fit. The second grant is madesubsequently in a fixed amount per acre of registered irrigated land within theWUA area.

The procedure for spending these funds is the following. WUA’s and theircommittees identify maintenance priorities and then ask special governmentunits to prepare plans and quantity estimates for the work to be done. Onceagreed upon by both sides, the government concludes a contract with theWUA to carry out the work. Then, the WUA can directly employ farmers orlocal artisans for this purpose or it can subcontract to a third party. Thegovernment then makes payments to the WUA on a predetermined schedulerelated to work accomplishment.

funds for maintenance to be very low (see Module No.3 of this Guide).This means that incentives are not fullyeffective.

Institutional Arrangement 4: Service provision incharge of a financially autonomous governmentagency (see Fig. M9-4).

In such cases, a government agency or parastatalorganization is the arranger and payer of maintenanceservices. Water users are consumers of these servicesinsofar as they can take advantage of the level of waterdelivery service which they have requested. Thegovernment agency or parastatal organization caneither perform the maintenance service on its own –“en régie”– or subcontract it to a commercial provider.Water fees paid by the users provide sufficient costrecovery.

In general, such arrangements with respect toincentive effectiveness can be rated as follows:

For the provider side : +For the user side: +

The agency holds both the arranging and thepayment function vis-à-vis the provider. In cases whereno subcontracting is foreseen these functions coincide

168

Module 9

Box M9-3: Service provision by financially autonomousgovernment agency

Arrangements like these are found in the above-mentioned region inSouthern France where the “Compagnie d’Aménagement des Coteaux deGascogne” (CACG) holds the mandate to manage water resources (seeMAINTAIN Case Study No. 2). The institutional arrangement discussed herecorresponds to the situation faced by CACG in the so-called “périmètres enconcession”. Here, the CACG is commissioned by the state to operate andmaintain the irrigation facilities in various perimeters totalling some 70,000ha. These services are paid for by the contractually agreed upon fees theirrigation farmers pay the CACG.

169


with the provider function (service provision ‘enrégie’). In both cases, there is no customer sovereigntyof the users toward the service provider.Consequently,the providing units will feel accountable to the agencyand not to the users, an arrangement which may giverise to services that do not really correspond to actualneeds. Incentives on the provider side hence are moreor less deficient, depending on the managementcapacity of the agency.

Incentive effectiveness on the user side is restrictedby the missing customer sovereignty.However,the loopof services and payments is closed due to the financialautonomy of the agency.The user fees directly feed intothe service provision process and thus provide asecondary link between payment and performance.Incentives on the user side to request and monitormaintenance services (and withhold payments in caseof unsatisfactory provision of services) may be low,butare still not without effect.

Institutional Arrangement 5: GovernmentAdministration

This is the well-known case of medium and large-scale government administered schemes, as they existin many parts of the world, especially in South Asia.

The feedback loop between service provision andpayments is interrupted. Maintenance budgets areallocated by Ministries of Finance without any obviouslink to service performance. There is no customersovereignty at all,since maintenance is either providedor subcontracted under the responsibility of particulargovernment departments. Water fees normally arehighly subsidized, however without any link of thesubsidy to variables that influence service qualiy.

In this case,the answers to all our three key questionspoint to low or non-existing incentive effectiveness,thelikely ratings in such cases will be:

For the provider side: -For the user side: -

170

Module 9

Figure M9-1: Private Service Provision

Figure M9-2: Government Subsidy to Private Service Provision

Figure M9-3: Government Subsidy and Split Arranging Function

171


Figure M9-4: Financially Autonomous Government Agency

Figure M9-5: Government Administration

Module 10

Actor Specific “Incentive Profiles”

Irrigation in most countries of the world faces aproblem which we have described in chapter 1 of PartOne of this Guide as a “maintenance paradox”. Whilepoor maintenance has been found to be one of themajor causes for suboptimal performance of irrigationsystems world wide, irrigation research and projectappraisals in most cases hardly pay any attention tomaintenance.

As we have stressed in the introductory chapters tothis MAINTAIN Guide,we adopt the premise here thatthe general lack of commitment toward solving theproblem of inadequate maintenance of irrigationsystems is primarily due to weak institutions that do notcreate appropriate incentives to motivate stakeholdersto ensure adequate maintenance.

To illustrate this point,we repeat in Table M 10-1 theexamples of incentive deficiencies that are prevalent ina conventional top-down administrative setting ofirrigation maintenance.

When talking about incentives,we need to be awareof the difference between incentives and motivation.An incentive is any reward or sanction, intended orinadvertent, which has as its effect a modification ofbehavior. In contrast, motivation is the condition ofbeing encouraged to behave in a certain manner.In other words,incentives,as rewards or sanctions,arecauses that bring about the effect of motivation.Theydo this if and only if appropriate conditions or

172

Module 10

Why is the analysis of

incentive deficiencies

essential in the

context of

maintenance service

provision?

The purpose of this Module is to provide guidance for the identification andelimination of incentive deficiencies in maintenance service provision forirrigation. In so doing, the Module refers to previous Modules as well.

What are “incentives”

and when are people

motivated to act in a

certain way?

173


“conditioning factors” are given. An increase in salarymay be forseen by the management of an irrigationagency to motivate its staff to tend more intensively tomaintenance issues. However, if the condition forsalary increase is the number of years in service (the“conditioning factor”), and if the value system of thestaff is dominated by monetary benefits, it is highly

Table M 10-1: Examples of incentive deficiencies for irrigationmaintenance

Stakeholder Incentive Deficiencies

Senior Irrigation Officials � Low political benefits, high opportunity costs

� Low, delayed visibility of benefits of maintenance� Low budget priority� Rehabilitation projects create political support

Irrigation Agency � Budget allocations unrelated to fee collection ratesManagement � Total control over O&M funds requires less

accountability to users than cost-sharing arrangements

� Accountability to internal hierarchy, not water users

Operational Staff � Maintenance lacks professional appealof Irrigation � Deterioration rewarded by rehabilitation projectsAgencies � Accountability to internal hierarchy, not water users

Water Users � Irrigation infrastructure seen as government property and responsibility of government to maintain

� No relation between payment of water fees and quantity or quality of maintenance

� No clear water rights� Not involved in priority setting for maintenance works

Foreign Donors � Difficulty monitoring use of resources for maintenance� Difficulty monitoring benefits of effective maintenance � Pressures to perpetuate financing of capital intensive

projects, such as rehabilitation, modernization and expansion

unlikely that any improvement in maintenanceprovision will come about with such an incentive.Unless maintenance activities and results are linked byan appropriate conditioning factor to the value systemof the agency staff (e.g.using a premium system that istied to the results of external audits) motivation effectswill remain negligible.

Most theories of motivation (e.g. the much-citedtheories of Maslow, Herzberg and others) assume thatpeople have needs for money, belonging, friendship,recognition and so on, to which they react. Hence, if a“manager”can identify these needs and manipulate themeans for their achievement, then his purposes can beserved as people respond to their internal drives.

Entering the world of service provision,as it has beenreferred to in this Guide,we have to realize that

� In service provision there are several – at least two –actors (“exchange partners”) that are independent inthe sense that they are not necessarily engaged in a“manager-subordinate”relationship with each other.Hence,the assumption,that there is a “manager”who“elaborates”an incentive system for the other actorsis an oversimplification in many cases.

� In service provision and especially in complexservice delivery systems, people interact with otherpeople.This means that the assumption that peopleare basically passive responders to externalincentives is also an oversimplification. Instead,people devise strategies to fulfill their needs andwants in a proactive way.

� Thinking in terms of service delivery emphasizes thenecessity for considering relationships betweenpeople in order to understand their behavior. Insystems of multiple actors, such relationships aresubject to different goals and expectations of theactors in the exchange. Hence, they create bothopportunities and threats for the involved actors.These are factors that influence motivation.

174

Module 10

Taking such a perspective and extending traditionalthinking on motivation into the realm of servicethinking and exchange relationships, we take thefollowing conceptual approach:To see people act in acertain way, it is not sufficient to have an attractiveincentive system with functioning “conditioningfactors”. Instead,people need to have:� A.The possibility to act in a certain way,meaning:

– (A1) Framework conditions must be such thatpeople are not constrained to act in the intendedway,

– (A2) People must have a chance to participate inthe process of determining the “intended way”ofacting (“goal setting”),

– (A3) Necessary interactions with other peopleneed to come about in such a way that agreementsabout content and conditions of the exchangerelationship are established and adhered to,

– (A4) People must receive appropriatecompensations for their actions/services and thusreceive the necessary resources or financial abilityto go on acting as desired or agreed.

� B.The ability to act in the desired way,meaning:– (B1) People must have the qualification or

functional ability to act in the desired or agreedway,

– (B2) People must have the social/politicalauthority to act in the desired way,

– (B3) People must have the appropriate rights to actin a certain way (e.g.use rights, control rights).

� C. The motivation to act in the intended way,meaning:– (C1) People must perceive the benefits of the

action that accrue to themselves (includingmaterial and immaterial incentives) to exceed thecosts (including transaction costs) of their owncontribution,

– (C2) People must have confidence in the terms andconditions of the agreements, rules,contracts or

175


176

Module 10

procedures that govern their interactions with otherstakeholders,– (C3) People must see the goal of the action in line

with their own interests and value system.In this Guide, we perceive “incentives” and the“incentive systems” not only in terms of rewards andsanctions but also in terms of the above-mentionedconditions that create both the motivation to act andthe possibility and ability to act.We assert that the lackof possibilities and/or abilities to act can be perceivedas serious disincentives by the potential actors.

This Guide and the entire set of Modules are acontribution to the analysis and improvement ofincentive deficiencies in irrigation maintenance. Withreference to the above- mentioned factors of acomprehensive incentive system, the Guide providesthe following concepts.� A.The possibility to act in a positive way with respect

to maintenance provision:– (A1) requires the analysis of “framework

conditions”7 and the tailoring of maintenancestrategies to the respective conditions. With thepresentation of the instrument of “StrategicInstitutional Positioning” (SIP) the Guideresponds to this requirement (see chapter 4.1).SIP provides direction for devising situation-specific maintenance strategies that fit with theexisting institutional framework conditions.

– (A2) demands that the major stakeholdersparticipate in determination of objectives.In Module1, this Guide presents approaches to determiningobjectives that explicitely consider the role of thestakeholders in the process of setting objectives.Moreover, it shows how such approaches need todiffer with varying framework conditions.

How does this Guide

contribute to

Incentive Deficiency

Analysis and to the

solution of incentive

problems in

maintenance

provision?

7 Essential elements of the framework conditions in a service relationshipare the “external institutional environment” (see chapter 3.1) and the“service agreement” (see chapter 3.2)

– (A3) asks for appropriate compensations formaintenance provision to be channeled to theservice providers.While this Guide does not touchupon the questions of water pricing and tariffsetting, it deals with the necessary institutionalarrangements for irrigation financing (Module9). Doing so, it discusses a topic which is one ofthe major causes for deficient maintenancefinancing.

� B.The ability to act in a positive way with respect tomaintenance provision requires that:– (B1) people must have the qualification or

functional ability to act in the desired way.We hopethat this Guide can help to improvequalifications of irrigation managers, plannersand stakeholder representatives to deal inappropriate ways with topics that are essentialfor effective maintenance provision.

– (B2 and B3) people must have the necessaryauthority and appropriate rights to act in apositive way with respect to maintenance. InModule 5, the Guide presents an analysis ofproperty rights and authority systems and thuscontributes to overcome respective deficiencies.

� C. The motivation to act in a positive way withrespect to maintenance provision demands that:– (C1) people perceive the benefits of the action that

accrue to themselves (including material andimmaterial incentives) to the costs (including thetransaction costs) of their own contribution.InModule 3, the Guide presents a method for a“Rapid Assessment of Economic Incentives” formaintenance provision.With such a tool it will bepossible to make a “quick and dirty” judgementas to the cost-benefit situation as it is perceivedby the involved stakeholders.

– (C2) people must have confidence in the terms andconditions of the agreements, rules, contracts orprocedures that govern their interactions withother stakeholders. In Module 6, the Guide

177


178

Module 10

introduces ways and means for how to analyseand improve the governance of maintenanceprovision.The tools presented in that Module areillustrated in numerous case examples of theMAINTAIN Case Studies.

– (C3) people must see the goal of the action to bein line with their own interests and value system(“goal congruence”).

As mentioned above,we perceive the actors involvedin maintenance provision not just as passive reactorsbut as active initiators who pursue personal interests8

and devise strategies to pursue these interests.Therefore, we need to take these interests and

strategies into account and find out whether or not andin what way these interests and strategies and themotivation to pursue them are effected by incentivesand disincentives.

An ASIP is a format that tries to give a summaryoverview of different incentives and disincentives facedby an individual actor in connection with a particularservice provision,in our case in connection with waterdelivery and/or maintenance in irrigation.Moreover,theASIP identifies how these incentives and disincentivesare linked to the actual or assumed interests andstrategies of this actor.Trying to assess the effect of thedifferent (dis)incentives on the motivation of the actorin question, ASIP provides a base for discussions onoverall “incentive effectiveness” of a given servicesystem relative to the specific actor. This overviewprovides a listing or profile of assessments of incentive

What is an “Actor-

Specific Incentive

Profile” (ASIP)?

In the following, we present the instrument of “Actor-Specific IncentiveProfiles” that directly relates to this latter requirement.

8 When we speck of the personal goals of individual actors – organiations,groups, persons – we use the term “interests” in order to make a cleardistinction to the official goals set by organizations and groups.

effectiveness of a given service system arrangement fora selected actor in that system.

The methods for implementing an ASIP may varydepending on the circumstances.There may be “quickand dirty”rapid assessments,where rough assumptionsabout a particular actor’s main interests and strategiesare made, without this actor or representatives of thisactor being present. There may be other situations,where in-depth workshop discussions seemworthwhile, with representatives of the actor inquestion playing a central role.

A particular difficulty is the identification of theinterests and strategies of an organization or a group.The interest or personal goals of such an organizationor group might differ substantially from its official goal.The official goal of a department may be to ensureoptimal maintenance of irrigation infrastructure.However one may assume that the interest of thedepartment may instead be to look after its own growthand survival and secure a rising share of next year’sbudget allocation. At the same time such a suggestionmay be a gross misjudgement of a loyal and engageddepartment head who does his or her best to do a goodmaintenance job in the most cost effective way.

This is why it is important to realize that ASIP doesnot aim to work with real but with assumed interests.ASIP’s intention is to check the robustness of incentiveseven in case the particular actor might have highlyopportunistic interests that partly run counter to aneffective and efficient service delivery. Incentiveeffectiveness in such a case will be achieved if sufficientrewards or sanctions are present in the existinggovernance system to make the actor realize thateffective and efficient service delivery will be in his/herown best interest. This is what is meant by creating a“win-win-situation”.The actor realizes that the best wayto serve one’s own interests is to serve the interests ofthe service provision agreement at hand.

179


Methods for

implementing an

“Actor Specific

Incentive Profile”

(ASIP)

180

Module 10

Box M10-1: Implementing ASIP – The case of the “NesteSystem” in southern France (from MAINTAIN Case Study No. 2)

The “Neste System” is a system of storage reservoirs, canals, small riversand irrigation schemes that is located in the region of Midi-Pyrénées, insouthern France. The hilly landscape in that region stretches from the southto the north and is interspersed with a total of 17 small rivers and streams,which, owing to the morphological structure, only have extremely smallwater-catchment areas. As a result, water flow would, under normalconditions, not be possible all year round. To improve the availability ofwater, both for agricultural purposes and for drinking water supplies to thecities and local communities in this area, a link canal (“canal de la Neste”)was built some time ago. This canal is fed by storage dams in the Pyreneesand carries water both to the river Neste and to the head of the other 17small rivers and thus makes it possible to provide minimum flows of watereven during the time of the year when these rivers would normally run dry.

The “Compagnie d’Aménagement des Coteaux de Gascogne” (CACG) is asemi-public company (“societé d’économie mixte”) that is mandated by thestate to promote water development in the region of Midi-Pyrénées. Amongstothers its mandate comprises the maintenance of the mentioned system ofthe “canal de la Neste”. The “client” of this maintenance service is the state,which wants to preserve the public infrastructure of the canal system for thefuture, but indirectly also the water users (for irrigation as well as municipaland industrial water supplies) who are the clients of various water deliveryservices. The maintenance service of the canal system is financed fromdifferent sources. First, are the water charges paid by the users. Second is asubsidy by the state to CACG for the service of maintaining minimum flowrates in the rivers. This subsidy is financed from a special water duty, which isa fee paid to the water agency by the users. The level of this subsidy,however, is linked to certain ideal hydrographic standards of water provisionin the canal system, which are monitored by the state. Moreover, a de-factosubsidy is given in that the state has been financing a large-scalerehabilitation program for the entire Neste canal system before CACG tookover the maintenance mandate.

The case is interesting for an ASIP exercise, an exercise carried out by theauthors of this Guide after completion of the MAINTAIN study and withoutfeedback from CACG. At the time of the study, CACG did not consider itselfto be facing any maintenance problems worth mentioning. Moreover, itclaimed to be one of the few large-scale, regional water managementorganizations that achieves full cost recovery in operating and maintaining itsvarious sub-systems. Moreover, CACG regards itself explicitly as a serviceprovider and has fully internalised the role. This was indicated clearly in itsvery positive incentive profile.

181


The essential steps needed to implement an ASIP arethe following:

� Identify the type of service in question,� Identify the actor, i.e. the organization, group or

individual for which the ASIP is to be implemented,� Discuss with representatives of the actor personal

goals and strategies. In case this is not possible,assume such personal goals and strategies pursuedby similar actors in similar situations,

Box M10-2: Implementing ASIP – The case of medium and largeirrigation systems in Pakistan and India (Source: adapted from Ul Hassan, 1999, MAINTAIN Country Study No.2)

Both in Pakistan and India, irrigation and drainage are crucial for theeconomy. Some of the world’s largest surface irrigation systems exist inPakistan and India. Following the Green Revolution of the 1960’s, manyirrigation systems were extended and storage dams were constructed toprovide increased and reliable water supplies. The major focus up to the1970’s had been to extend the physical infrastructure and few attempts weremade to sustain the available facilities. The financial resources had beendiverted to extend irrigation infrastructure, rather than to manage andmaintain. Owing to increasing costs of new irrigation projects and increasingscarcity of water resources, in Pakistan the 1980’s witnessed a shift inemphasis to the conservation of water through lining of tertiary canals andrehabilitating the primary and secondary canal systems through severalprograms.

However, in both countries, water delivery remained subsidized with no directlink between service and costs. Water-tax collection has tended to declineover time. The institutional environment fostered poor coordination andresulted in rendering official rules for water delivery and maintenance serviceprovision impossible to enforce. System inefficiencies have led to variousrecent institutional reforms in the water sector. An example of one prominentand recent reform process is Andhra Pradesh, India (analysed with respect toservice provision in MAINTAIN Case Study No.5).

The ASIP method employed in Case Study 5 is only a rapid assessment ofthe incentive situation of irrigation staff as it may have existed in manyirrigation schemes prior to the reform processes. The example shows, that,from the standpoint of incentive effectiveness, there was no chance for areasonable functioning of the maintenance service provision in theseschemes.

182

Module 10

Tab

le M

10-2

. Act

or-

Sp

ecifi

c In

cent

ive

Pro

file

(AS

IP) f

or

mai

nten

ance

ser

vice

pro

visi

on

by

CA

CG

/Fra

nce

(Sup

por

ting

Doc

umen

t: M

AIN

TAIN

Cas

e S

tud

y N

o.2)

Typ

e o

f S

ervi

ce:M

aint

aini

ng t

he h

ydra

ulic

infr

astr

uctu

re in

the

Nes

te C

anal

Sys

tem

, Gas

cogn

e, F

ranc

e

Ince

ntiv

es:p

rovi

ded

by

the

Ser

vice

Arr

ange

men

t C

AC

G –

Gov

ernm

ent

Act

or

for

who

m in

cent

ives

are

ana

lyse

d:C

AC

G/F

ranc

e

“Per

sona

l”

Typ

e o

f m

oti

vati

on

Typ

e o

f in

cent

ives

Co

ndit

ioni

ng f

acto

rE

ffec

tive

ness

of

go

al/s

trat

egy

or

dis

ince

ntiv

es

ince

ntiv

es f

or

hig

ho

f ac

tor

pro

vid

ed b

y se

rvic

e ag

reem

ent

qua

lity

mai

nten

ance

Sec

ure

finan

cial

�

Mat

eria

l inc

enti

ves

“hea

lth”

of

CA

CG

Min

imiz

e in

itial

cos

t of

Im

ple

men

tatio

n of

sp

ecia

l C

omp

letio

n of

reh

abili

tatio

nhi

ghm

aint

enan

ce fo

r N

este

ex

-ant

e re

hab

ilita

tion

pro

gram

prio

r to

tra

nsfe

r of

mai

nten

ance

Can

al S

yste

mb

y go

vern

men

tre

spon

sib

ility

to

CA

CG

Ear

n m

aint

enan

ce

Use

r ch

arge

s fo

r su

pp

ly o

fU

ser

char

ges

to b

e m

ade

high

rese

rves

(“p

rovi

sion

de

wat

er in

wat

er c

ours

es t

oav

aila

ble

to

CA

CG

mai

nten

ance

”)b

e us

ed fo

r re

serv

e

Ear

n ad

diti

onal

sub

sid

ies

Pro

vid

e su

bsi

dy

for

mai

ntai

ning

Le

vel o

f sub

sid

y lin

ked

to

high

min

imum

flow

in r

iver

shy

dro

grap

hic

stan

dar

ds

of w

ater

pro

visi

on

Pot

entia

l use

of

Pre

scrip

tion

of m

inim

umS

ettin

g an

nual

mai

nten

ance

fairl

y hi

ghm

aint

enan

ce fu

nds

exp

end

iture

for

mai

nten

ance

rese

rve

for

spec

ific

low

er

for

othe

r ac

tiviti

esb

y go

vern

men

t (d

isin

cent

ive)

limit

of e

xpen

ditu

res

�Im

mat

eria

l inc

enti

ves

Sec

ure

op

erat

iona

l C

ond

uct

good

A

ct a

s st

ate

conc

essi

onai

reM

onito

ring

by

the

stat

e vi

ahi

ghin

dep

end

ence

m

aint

enan

ce b

ut in

ow

n“C

AC

G s

uper

viso

ry b

oard

“an

d h

igh

pre

stig

e p

lann

ing

and

imp

le-

for

CA

CG

men

tatio

n re

spon

sib

ility

Tem

pta

tion

to d

o “g

old

P

resc

riptio

n of

max

imum

Set

ting

of a

nnua

l mai

nten

ance

fairl

y hi

ghp

late

d“

mai

nten

ance

exp

end

iture

for

mai

nten

ance

re

serv

e fo

r sp

ecifi

c up

per

lim

it (d

isin

cent

ive)

of e

xpen

ditu

res

183


� Identify the types of motivation envisaged to resultfrom incentives,

� Identify existing or envisaged material andimmaterial incentives,

� Identify the conditioning factors that link particularincentives to the envisaged types of motivation,

� Asses the “incentive effectiveness”, i.e. the estimateddegree to which the identified incentives bring aboutthe envisaged motivation.

In the following we present two examples forimplementing ASIP, both of which are of the rapidappraisal type mentioned above. However they aremeant to give a clear orientation for how to go abouteven in more detailed efforts to elaborate ASIP formats.The examples refer to cases that are described in therelevant MAINTAIN documents mentioned in the titleof the figures. For those readers that do not have theopportunity to look through these papers, we givesome short indications as to the services in question.

184

Module 10

Tab

le M

10-3

Act

or

Sp

ecifi

c In

cent

ive

Pro

file

(AS

IP) f

or

mai

nten

ance

ser

vice

pro

visi

on

in m

any

Ind

ian/

Pak

ista

ni ir

rig

atio

n sc

hem

es

(Sup

por

ting

doc

umen

ts: M

AIN

TAIN

Cou

ntry

Stu

dy

No.

1, M

AIN

TAIN

Cas

e S

tud

y N

o.5)

Typ

e o

f S

ervi

ce: M

aint

aini

ng t

he h

ydra

ulic

infr

astr

uctu

re in

age

ncy-

farm

er m

anag

ed ir

rigat

ion

sche

mes

in In

dia

/Pak

ista

n.In

cent

ives

: Pro

vid

ed b

y go

vern

men

t em

plo

ymen

t co

nditi

ons

and

wor

k ru

les.

A

cto

rs f

or

who

m in

cent

ives

are

ana

lyse

d:E

ngin

eerin

g st

aff o

f Irr

igat

ion

Dep

artm

ents

/ Ir

rigat

ion

and

Com

man

d A

rea

Dev

elop

men

t D

epar

tmen

ts

“Per

sona

l” g

oal

/Ty

pe

of

mo

tiva

tio

nTy

pe

of

off

icia

l C

ond

itio

ning

fac

tor

Eff

ecti

vene

ssst

rate

gy

of

acto

r(d

is-)

ince

ntiv

es

of

ince

ntiv

es

Incr

ease

mat

eria

l M

ater

ial

wel

l bei

ng

ince

ntiv

eso

f se

lf/fa

mily

gain

Sal

ary

Leve

l of s

alar

y lin

ked

to

leve

l of

low

mon

etar

y in

com

ep

ositi

on, n

ot t

o q

ualit

y of

ser

vice

Imp

rove

mon

etar

y P

rom

otio

nP

rom

otio

n d

epen

dan

t on

exa

min

atio

ns/

low

inco

me

qua

lific

atio

ns/y

ears

of s

ervi

ce a

nd p

oliti

cal

conn

ectio

ns, n

ot o

n q

ualit

y of

ser

vice

Tem

pta

tion

to e

xtra

ct

Forb

id r

ent

Low

leve

l of t

rans

par

ency

and

very

low

sid

e p

aym

ents

for

oper

atio

nse

ekin

g ac

tiviti

esac

coun

tab

ility

rel

ated

to

serv

ices

and

mai

nten

ance

of

wat

er d

eliv

ery

and

mai

nten

ance

.se

rvic

es fr

om fa

rmer

sLe

ads

to r

ent

seek

ing

pos

sib

ilitie

san

d d

ecre

ases

qua

lity

of s

ervi

ce

Tem

pta

tion

to e

xtra

ct s

ide

Forb

id r

ent

Low

tra

nsp

aren

cy o

f ten

der

ing

and

low

pay

men

ts fo

r m

aint

enan

ce /

se

ekin

g ac

tiviti

es

cont

ract

ing.

No

exte

rnal

con

trol

/aud

iting

reha

bili

tatio

n co

ntra

cts

Gai

n m

ater

ial o

r G

ain

don

or fu

nds

with

Don

or fu

ndin

g on

ly fo

r re

hab

ilita

tion,

lo

wp

oliti

cal r

ewar

ds

reha

bili

tatio

n ne

eds

not

for

mai

nten

ance

Imm

ater

ial i

ncen

tive

s

Gai

n st

atus

and

P

refe

r “h

igh

stat

us“

engi

neer

ing

No

or lo

w s

anct

ioni

ngN

o vo

ice

for

nega

tivel

y ef

fect

ed fa

rmer

slo

wp

rest

ige

fro

m w

ork

wor

k (d

esig

n, c

onst

ruct

ion)

of d

efer

red

mai

nten

ance

Pre

fer

“hig

h st

atus

” P

rom

otio

n w

ith

Hig

h q

ualit

y m

aint

enan

ce s

ervi

ce r

equi

res

low

wor

k lo

catio

n in

tow

nstr

ansf

er t

o to

wn

freq

uent

pre

senc

e in

the

fiel

d

185

References and Literature


A. Main references: The MAINTAIN-papers

Series of case studies

No.1 Institutional Analysis of Maintenance in the Lower SeyhanIrrigation System, Turkey. Waltina Scheumann and ArturVallentin,1999

No.2 Maintenance as a Service Provision in Irrigation – TheExample of the “Neste System”in Southern France –. WalterHuppert and Christian Hagen,1999

No.3 Institutional Analysis of Water Delivery and MaintenanceService Provision in Irrigation:The Example of the JordanValley.Walter Huppert and Klaus Urban,1999

No.4 Shifting Public and Private Roles in Maintenance ServiceProvision:California’s Central Valley.Walter Huppert,MarkSvendsen and David Cone,2000

No.5 Incentive Creation for Irrigation System Maintenance andWater Delivery: The Case of Recent Reforms in AndhraPradesh. Mark Svendsen and Walter Huppert, 2000

No.6 Institutional Analysis of Maintenance Service Provision inthe Alto R’o Lerma Irrigation District, Mexico. Klaus Urban,Philippus Wester and Wim H. Kloezen, 2000 (CooperationGTZ – IMWI)

No.7 The Case of a New Water Fee System in the Republic ofMacedonia. Thilo Hatzius,2000

Series of thematic papers

No.1 Irrigation Maintenance Strategies:A Review of the Issues.Ian Carruthers † and Jamie Morrison,1994

No.2 The Neglect of Maintenance in Project Planning:What isthe Scope for Fine Tuning Appraisal Procedures? IanCarruthers † and Jamie Morrison,1994

No.3 Improving Maintenance in Irrigation: External Roles inStrategic and Operational Change.Mark Svendsen,1994

No.4 Maintenance, Sustainability, and Irrigation ManagementTransfer.Mark Svendsen, Ian Carruthers † and Thomas Buhl-Böhnert, 1994

186


No.5 Institutional Reform Options for Overcoming theUnderfinancing of Irrigation Maintenance in LessDeveloped Countries.Douglas L.Vermillion,1999

No.6 Property Rights and Maintenance of Irrigation Systems.Ruth Meinzen-Dick,2000 (Cooperation GTZ – IFPRI)

No.7 Irrigation Management Transfer: Changing ComplexDelivery Systems for O&M Services.Walter Huppert, 1997

No.8 Using Asset Management Techniques for Condition andPerformance Assessment of Irrigation and DrainageInfrastructure.Martin Burton,2000

No.9 Rent seeking in irrigated agriculture: institutionalproblem areas in operation and maintenance. JochenRenger and Birgitta Wolff, 2000

No.10Governance by Contractual Rules – Improving ServiceRelations in Irrigation. Birgitta Wolf and Walter Huppert,2000

No.11Optimal Maintenance in Irrigation.(4 Papers) MarkSvendsen and Walter Huppert; Gilbert Levine; Klaus Urban;David G.Cone,2000

No.12 Incentives for Financing the Maintenance of Irrigationand Drainage Systems. Paul J.M. van Hofwegen, 2000(Cooperation GTZ – IHE)

Series of country papers

No.1 Maintenance in Turkish Irrigation and Drainage Systems.Waltina Scheumann,1999

No.2 Maintenance in Pakistani Irrigation and Drainage Systems.Mehmood Ul Hassan,1999 (Cooperation GTZ-IWMI)

Accompanying GTZ-publication No. 263

Huppert, W. and K. Urban, 1998 Analysing Service Provision –Instruments for development cooperation illustrated byexamples from irrigation, GTZ

187


B. Additional references

Bromley,D.W.,1991 Environment and economy.Property rights andpublic policy.Cambridge:Blackwell.

Burton, M. and R.P. Hall, 1999 Asset Management – Addressing theissue of serviceability. Irrigation and Drainage Systems, KluwerAcademic Publishers,November

Cornish,G. and J. Skutsch,1997 A Procedure for PlanningIrrigation Scheme Rehabilitation. Report OD/TN 84, HRWallingford,Wallingford,UK.

El-Askari, K.M.S., 2000 A Methodology for Expenditure Planningof Irrigation Infrastructure Using Hydraulic ModellingTechniques.Unpublished Thesis,University of Southampton,UK.

FAO, 1993 The State of Food and Agriculture;Water Policies andAgriculture Rome.

GICC, 1998 Guidelines for Irrigation Canal Control. Sir WilliamHalcrow and Partners,Swindon,UK and Institute of Irrigation andDevelopment Studies,University of Southampton,UK

Glennie, E.B., P. Timbrell and J.A. Cole, 1991 Manual of conditionassessment for flood defences.WRc Engineering, Swindon,UK.

Herder-Dorneich, Ph., 1983 Der Sozialstaat in derRationalitätenfalle Kohlhammer

Herder-Dorneich, Ph., 1986 Theorie der Sozialen SteuerungNomos Verlagsgesellschaft,Baden-Baden

Huppert,W.,2000 Governing Irrigation Service Delivery in WaterScarce Situations ICID Journal,Vol 49,No.1

Huppert,W.and K.Urban,1993 Service Analyses in Water and SoilAssociations – Taking the Nienburg Weser Association as anExample In: Zeitschrift für Bewässerungswirtschaft,Vol 28, No.2,1993

IIS 1995 Asset management procedures for irrigation schemes –Final Report. Institute of Irrigation Studies, University ofSouthampton, Southampton,UK.

Karunasena, H., 1993 Mobilizing Farmers’ Resources forMaintenance of Irrigation Schemes In:Papers presented to theAsian Regional Symposium held in Beijing, PRC, 24.27 May. HRWallingford,Wallingford,UK.

Krause, D. and V. Temple, 1988 Maintenance Management In:Water Systems Operation and Maintenance Workshop, SessionNotes.Denver,CO.

188


Mody,A., 1996 Infrastructure Delivery EDI Development Studies,The World Bank

Olson,M.,1968 The Logic of Collective Action New York:SchokenBooks

Ostrom,E.L.Schroeder and S.Wynne,1993 Institutional Incentivesand Sustainable Development Westview Press

Richter, R. and E. Furubotn, 1996 Neue Institutionenökonomik –Eine Einführung und kritische Würdigung J.C.B. Mohr (PaulSiebeck) Tübingen

Savas,E.S., 1990 Privatization : the key to better government.Chatham House Publishers,Chatham,N.J.

Skutsch, J.C., 1998 Maintaining the Value of Irrigation andDrainage Projects, Report OD/TN 90, HR Wallingford,Wallingford,UK.

Svendsen, M., 2000 Irrigation Management Institutions in theWestern United States:an Overview. Mimeo.Paper prepared forthe World Bank Institute

Thoreson, B.P., D.C. Slack, R.P. Satyal and R.S.S. Neupane, 1997Performance-based Maintenance for Irrigation Systems Journalof Irrigation and Drainage Engineering 123 (2):100-105

UNEP,1992 Saving our Planet Nairobi.Wolff, Birgitta, 1995 Organisation durch Verträge. Deutscher

Universitätsverlag.World Bank, 1994a A Review of World Bank Experiences in

Irrigation.Report No.13676,WashingtonWorld Bank, 1994b Governance: The World Bank’s Experience.

Report No.12094,Washington

C. Recommended further reading

Bandaragoda,D.J.,2000 A Framework for Institutional Analysis forWater Resources Management in a River Basin Context. IWMIWorking Paper no.5

Bandaragoda,D.J.,1998 Need for Institutional Impact Assessmentin Planning Irrigation System Modernization IWMI,Colombo,Sri Lanka

Bruns, B.R. and R. Meinzen-Dick, 2000 (eds.) Negotiating WaterRights, IFPRI,Vistaar Publications,New Delhi

Dinar, A. and A. Subramanian, 1997 Water Pricing ExperiencesWorld Bank Technical Paper No.386

Herder-Dorneich, Ph., 1986 Theorie der Sozialen SteuerungNomos Verlagsgesellschaft,Baden-Baden

Huppert,W.and K.Urban,1998 Analysing Service Provision –Instruments for development cooperation illustrated byexamples from irrigation GTZ Publication Series No.263

Malano, H.M. and P.J.M. van Hofwegen, 1999 Management ofIrrigation and Drainage Systems – A Service Approach IHEMonograph 3,Balkema Publishers Rotterdam/Brookfield

Murray-Rust, D. and W.B. Snellen, 1993 Irrigation SystemPerformance – Assessment and Diagnosis IIMI, Colombo, SriLanka

Olson,M.,1968 The Logic of Collective Action New York:SchokenBooks

Ostrom, E. L., 1990 Governing the Commons – The Evolution ofInstitutions for Collective Action Cambridge University Press

Ostrom, E.L., 1992 Crafting Institutions for Self-GoverningIrrigation Systems. San Francisco, Calif.: Institute forContemporary Studies Press.

Ostrom,E.L.Schroeder and S.Wynne,1993 Institutional Incentivesand Sustainable Development Westview Press

Pratt, J.W. and R.J. Zeckhauser, 1991 Principles and Agents: TheStructure of Business Harvard Business School Press

Richter,R. and E. Furubotn,1996 Neue Institutionenökonomik –Eine Einführung und kritische Würdigung J.C.B. Mohr (PaulSiebeck) Tübingen

Savas,E.S., 1990 Privatization : the key to better government.Chatham House Publishers,Chatham,N.J.Skutsch, J.C.,1998 Maintaining the Value of Irrigation and Drain-

age Projects, Report OD/TN 90,HR Wallingford,Wallingford,UK.Subramanian, A., N.V. Jagannathan, R. Meinzen-Dick, 1997 User

Organizations for Suistainable Water Services World BankTechnical Paper No.354

Svendsen,M.,2000 Basin Management in a Mature Closed Basin:The Case of California’s Central Valley.Case Study prepared forthe International Workshop on Integrated Water Management inWater Stressed River Basins Loskop Dam, South Africa, 16–21Oct.

189


Vermillion, D.L. and J.A. Sagardoy, 1999 Transfer of IrrigationManagement Services – Guidelines FAO,Rome

Wolff, Birgitta, 1995 Organisation durch Verträge. DeutscherUniversitätsverlag.

World Bank, 1994b Governance: The World Bank’s Experience.Report No.12094,Washington

190


Authors’ addresses

Prof.Dr.Martin Burton Consultant,Spring Cottage, Itchen Stoke,United [email protected]

Dr.Paul J.M.van Hofwegen Senior LecturerIHE-Delft,The [email protected]

Dr.Walter Huppert,Senior Planning Officer,GTZ,Eschborn,[email protected]

Dr.Ruth Meinzen-Dick,Senior Research FellowInternational Food Policy Research Institute,Washington D.C.,[email protected]

Dr.Waltina Scheumann,Assistant Professor,Technical University of Berlin,Institute for Management in Environmental Planning (IMEP) [email protected]

Dr.Mark Svendsen,Consultant,Philomath,OR,[email protected]

Dr.Klaus Urban,Consultant,Wiesbaden,[email protected]

191


Dr.Douglas L.VermillionConsultant,Spokane,WA,[email protected]

Prof.Dr.Birgitta WolffChair of International Management,Department of EconomicsOtto-von-Guericke-Universität,Magdeburg,[email protected]

192


Governing Maintenance Provision in Irrigation2001)_Governing...Ringler (Germany),Mark Rosegrant...

Documents

Transcript of Governing Maintenance Provision in Irrigation2001)_Governing...Ringler (Germany),Mark Rosegrant...