Accounting for water : institutional viability and impacts of market-oriented irrigation interventions in Central Mexico

During the past decade, many countries throughout the world have attempted to improve their generally poor performance record of agency-managed irrigation systems by designing and implementing institutional policy programs. This thesis analyses the institutional viability and the local impact on irrigation performance of two such institutional intervention programs. This is done in the context of the Alto Río Lerma irrigation district (ARLID), a large-scale irrigation system with a command area of more than 112,000 hectares located in the State of Guanajuato, Central Mexico. The central notion that runs through this study is the recognition that new institutional arrangements do not necessarily follow institutional design principles and top-down directives from policy makers and government interveners, but rather are created in a process of local negotiation between water users, farmer leaders and irrigation managers. This requires detailed observations on local practices, strategies and interactions on how users, leaders and managers cope with building and transforming irrigation institutions.The first intervention program analyzed in this thesis aims to transfer management responsibilities from the irrigation agency (CNA) to local newly established water user organizations (WUAs) and to make them responsible for cost-recovery. The neoliberal belief behind this program is that greater farmer involvement in irrigation management and financing would eventually lead to a reduction in public investments in irrigated agriculture as well as to higher levels of irrigation performance. The second institutional intervention program that is analyzed in this thesis deals with the introduction of water markets, aiming at re-allocating water use to uses of higher economic value. Both intervention programs clearly embark on the 'less state, more market' approach to design and implement institutional arrangements for more use and cost-effective management of scarce resources such as water and funds. The central theme of this thesis is to study how water users and local water managers deal with the problems that result from these interventions and arrangements.In 1992, ARLID was one of the first of Mexican districts were irrigation management transfer (IMT) was introduced. The Mexican IMT program is worldwide considered to be the most successful and ambitious institutional programs in its kind, both because of its scale and the speed with which the policy was designed and implemented. The claims on the successes of these and similar programs are generally ideologically driven, and empirical evidence to support the policy claim that these institutional interventions have achieved these goals remain largely undocumented. This thesis fills some of this empirical void.This thesis comprises ten chapters. After the conceptual and methodological introduction in chapter 1, two contextual chapters follow. Chapter 2 introduces the research site, and chapter 3 presents the political economic history and context in which the institutional interventions took place. Chapter 4 to 9 provide the empirical data from the intensive case study in ARLID. Chapter 10 evaluates the main empirical, conceptual and methodological findings and discusses possible implications for policy and research orientations.The first chapter provides a review of the conceptual approaches and methodologies used to study the viability of new institutional arrangements in the context of market-oriented reforms. This thesis explores the kind of conditions, processes and mechanisms that help to create viable institutions. These mechanisms do not only refer to institutional and economic relationships, but also to the relationship between institutions, and water scarcity and irrigation technology. In that sense the thesis follows the notion that irrigation technology is a sociotechnical phenomenon, with institutional requirements for use.A key concept used for analyzing the viability of new irrigation institutions is accountability. This study shows that four forms of accountability can be separated: operational, financial, political and administrative accountability. Studying these forms of accountability (and the relationship between them ) demonstrates that practices and strategies of decision making in water management are only partly influenced by rational choice and (economic) incentives for utility maximization. They are also influenced by socio-political factors.The methodological concern of this thesis is to develop a set of tools and indicators that help to assess the impacts of institutional interventions, both in terms of changes in irrigation performance levels, and in qualitative terms of processes of institutional viability. Two sets of performance indicators are presented. The first set comprises indicators that measure actual performance against set performance targets. The second set compares performance levels over time and across different system levels. The field work for this thesis comprised a wide set of methods and research techniques, ranging from intensive observations of practices and strategies for coping with the new institutional arrangements, to daily measurements of water flows in two selected WUA areas and several canals and fields in the irrigation district. Field research took place during the period from early 1995 to mid-1998. Chapter 2 provides background information on the location, history, water availability and layout of ARLID. It also introduces the two main groups of farmers that now jointly are organized in the new WUA: ejidatarios and private growers. It shows how these two groups have historically and economically been separated for almost 70 years. The chapter also provides a brief description of the Cortazar and Salvatierra module areas, their canals and WUAs that were selected for the intensive case study.Two important notions form the basis of the chapter 3, in which a detailed description of the historical and political economic context of IMT and water marketing is given. The first notion is that the legitimization for institutional interventions should be viewed in the political economic context of privatizing the Mexican country-side through dramatic constitutional revisions in the way distribution of land and water rights are organized. These reforms and revisions were started in the 1980 as a result of both a political and an economic crisis and were accompanied by wave of programs on the liberalization, decentralization and technocratization of the economy. Institutional intervention programs like the Mexican IMT program do not come on their own, but are preceded, accompanied and followed by other intervention programs that share the same neoliberal legitimization for restructuring the political and economic control over resources. The second notion is that to achieve the economic goals of these neolibral programs, the Salinas and subsequent administrations needed intervention policies such as IMT and water marketing as these would help to recover costs from the users. However, to improve greater involvement of farmers in water management, neoliberal programs are not a necessity.Chapter 4 shows how IMT and water markets were implemented, both at the national level and in ARLID. The IMT program is a typical example of institutional engineering, comprising detailed prescriptions on institutional design and a top-down approach for implementing these designs. Yet, farmers and local managers do not fully disregard these proposed arrangements. Rather than using them as prescriptive dictates, the new principles and arrangements served as a strategic starting point for the WUAs for negotiating irrigation fee levels, water marketing, mode of user representation, maintenance programs, and redistribution of O&M responsibilities between the WUAs and CNA. Local recipients have partly adopted and accommodated the new institutional arrangements. The IMT program allowed local irrigation managers, new leaders of WUAs and farmers to jointly create room and institutional flexibility to alter the proposed arrangements. This was made possible mainly because from the initial stage of implementation, CNA officials in ARLID have shown to users that they are credible and willing to share knowledge and information in a transparent way. Credibility, transparency, institutional flexibility and probity were among the most important factors to explain how operational accountability from CNA to WUAs, and from WUAs to water users, was earned in a continuous process of negotiating institutional arrangements. Furthermore, given the wider set of market-oriented reforms and constitutional revisions that farmers had become familiar with, they realized that they could hardly resist IMT.Chapter 5 provides a detailed analysis of practices, results and operational accountability for local water management. It demonstrates that official post-transfer arrangements for O&M in ARLID do not always reflect actual irrigation practices. As result, outputs of these practices in terms of relative water supply and land and water productivity remained unchanged after IMT. The main reason for this is the conflict in forms of accountability, which could not be resolved under the new institutional arrangements. Rather than adhering to new principles of using water as an economic good, farmers and local managers adhered to pre-IMT arrangements and practices of distributing water. Daily measurements of water flows and observation of irrigation practices revealed that ditch tenders (who are now employed by the WUAs) generally report planned water deliveries rather than actually supplied volumes. In that sense, they continued the practice of being administratively accountable to both users and CNA. This runs counter to the idea of encouraging efficient water use through water pricing and cost-recovery and hence being both financially and operationally accountable. Comparison of water use per hectare or per m 3of water across the 11 module areas in ARLID, shows that although water is the limiting factor in ARLID, the irrigation management strategy chosen by farmers is to maximize returns per hectare rather than increasing returns per m 3of water. As a result, water deliveries per hectare of irrigated land are generally very high. The chapter also provides some evidence that existing irrigation technology and water management regimes do not always match with the institutional capacity of the new WUAs to operate the system. The chapter also explores some ideas of approaching issues of relative water scarcity in the context of conjunctive use and in a wider setting of the entire water basin.Chapter 6 gives a detailed account of how practices, processes and mechanisms of user representation, leadership election and conflict resolution affect political accountability and institutional viability. Detailed analysis of election procedures and processes over a period of two WUA administrations demonstrate that the de jure distribution of farmer representation in the boards and assemblies of WUAs is well organized. However, the de facto control over decision making remains in the hands of a few farmer leaders. These leaders have developed several strategies to reproduce this control. These include political coercion to become re-elected; favoritism in the employment of board members and irrigation staff; corruption; and extending economic and political networks. Particularly in the case of Salvatierra WUA, these strategies have caused that accountability was hardly created. Farmers use two strategies in their trying to counter-balance these mechanisms that weaken accountability. First, they refuse to pay higher fees to signal their dissatisfaction with th WUA's management policies. Second, they call for CNA to resolve disputes over illicit payments and the performance of the WUA leadership.In chapter 7 an analysis is made of how mechanisms of financial accountability influence the institutional viability of the post-transfer institutional arrangements for irrigation management. WUAs in ARLID succeeded to boost cost-recovery and to become financially autonomous. In that sense IMT has been very instrumental to achieving the government goal of reducing public investments in irrigation. However, improved cost-recovery and financial autonomy have also helped WUAs to create room for negotiating the reformulation of important institutional arrangements such as the percentages of total revenue that had to be paid to CNA; the level of water pricing; and the conditions and prices for water trading. Financial transparency, the use of computers and external auditing helped the WUAs to become financially accountable to fee payers. Moreover, particularly in the case of Cortazar, leaders showed that they were credible when they started to use revenues from fee payments to dramatically improve canal maintenance. They also showed their willingness to fire irrigation staff that had proved to be corrupt. Yet, financial viability is threatened in two ways. First, as a result of lack of political accountability (such as in the case of Salvatierra). Second, because of the WUAs' total dependency on water availability to generate income.In chapter 8 the focus of analysis is shifted towards issues of water marketing. WUAs in ARLID indeed have used the now legal opportunity to market water with other WUAs within the district. However, unlike the common belief among neoliberal advocates of water marketing, trading is not used to price water at opportunity cost levels. Rather, the process of water marketing is used to create solidarity among the WUAs. This solidarity is needed to be able to work together in the new federation of WUAs, which has opted to take over from CNA maintenance responsibilities for the main canal system. Water trading is also used by WUAs to show their members that they are credible in the sense that they will try to deliver water at any time; even when they have to buy it from other WUAs during times of water scarcity. As a result, prices of traded water are generally lower than water prices paid for normal water deliveries during times of sufficient water availability.Chapter 9 returns to the official objectives of the IMT program and assesses the program's impact on water management. By applying a set of comparative indicators, it is shown that the result of IMT and water marketing in terms of changes in performance levels are disappointing. With the exception of improvements in cost-recovery and maintenance, there is no evidence that water is used more efficiently or that IMT has resulted in higher levels of productivity per hectare or per m 3of water.Chapter 10 first provides a full synthesis of the empirical findings discussed bove. It subsequently revisits some of the conceptual and methodological tools used for this study. This shows that the conceptual notion of social requirements for use of irrigation technology should be further elaborated to institutional requirements for use. An important notion in this respect is that institutional arrangements are a mixture of formal prescriptive arrangements, existing arrangements and newly negotiated arrangements. Existing and new institutions are the result of a continuous process of creating, negotiating, earning, and maintaining accountability. This explains why institutions manifest themselves as being contingent rather than proposed static structures that follow designed institutional principles.A revisit of the accountability concept shows that it is useful to separate different forms of accountability. This separation helps to analyze accountability mechanisms that go beyond those related to providing mere O&M services or cost-recovery; it also helps to unravel in what way providers of services are accountable, or not; and it helps to recognize in what way different forms of accountability can compete with each other. This is particularly of interest in the case of joint WUA-agency management of irrigation system, where different organizations can have competing management values and targets, and hence try to be accountable in different ways.The methodological revisit on the different sets of impact assessment tools shows that comparative indicators are useful to assess changes over time as a result from institutional interventions. Yet, they are limited in the sense that they exclusively focus on economic and productivity output oriented results. Furthermore, applying these indicators was more time and resource consuming than assumed. The reason for this are the complexity and amount of secondary data needed as well as the difficult process it takes to deconstruct the meaning of these data.This study has several implications for market-oriented policies, both in and outside of Mexico. Because of its unique political economic and constitutional context, the 'Mexican IMT model' cannot simply be copied to other countries where similar attempts of planned institutional changes are underway. Particularly the fact that IMT and water marketing in Mexico were supported by political commitment at the highest levels and by dramatic constitutional revisions and other agrarian reforms, explains why the new institutional arrangements for water management could be speedily implemented on such a wide scale. This study also suggests that institutional 'modernization' programs (such as IMT and water marketing) and physical modernization programs in irrigation should be better matched. Although physical improvement programs are often used to sell institutional reforms (and the consequent burden) to water users, they hardly ever are matched in such a sense that the proposed new technology matches with the proposed new institutional arrangements to use this technology.Chapter 10 ends with a discussion on why to orient research on institutional interventions in water management to the level of entire water basin rather than irrigation systems. Also, irrespective that universities and international research institutes need to do research in the context of a globalizing world and with globalized research agendas, a plea is made to not shift too frequently to doing extensive, quantitative and comparative studies during a short period of time. Understanding institutional change processes and their impacts at the local level requires intensive and qualitative case studies over longer periods of time. Only in this way collaboration with local research partners can fully develop, resulting in the level of creation of institutional research memory that is needed for this understanding