Spanish Water Policy and the National Hydrological Plan:

An Advocacy Coalition Approach to Policy Change











Jeanie Bukowski

Institute of International Studies

Bradley University











Prepared for presentation at the EUSA Ninth Biennial International Conference, Austin, Texas, March 31-April 2, 2005.









DRAFT – not for citation without permission of the author. Comments and suggestions welcomed.


The policy networks approach to understanding governance in modern societies is increasingly gaining currency among scholars who conceptualize the European Union (EU) as a multi-level system. For many analysts, the idea of intricate webs of public and private actors at various levels connected by common interests and dependence on resources would seem to capture the complexities of policy making better than other approaches.

Policy network analysis is not without its critics, however. Among the criticisms leveled at this approach are that: 1) it is largely a descriptive model with few theoretical implications; 2) it underestimates the difficulty of delineating policy networks, particularly within the European Union; 3) it suffers from lack of definitional clarity; and 4) gathering empirical data within the rubric of policy networks is time consuming and does not provide sufficient yield (in terms of generalizable results) for the effort.

The current paper seeks to address these criticisms by exploring a particular approach that falls generally within policy network analysis, the advocacy coalition framework (ACF) developed by Sabatier and Jenkins-Smith (Sabatier 1987, 1988, 1991; Sabatier and Jenkins-Smith 1993). The ACF defines advocacy coalitions as consisting of public and private actors at all levels of government who have in common a set of  beliefs (policy goals coupled with causal and other perceptions) (Jenkins-Smith and Sabatier 1993: 5). The framework views policy change as a function of: the interaction of competing advocacy coalitions attempting to translate their belief systems into governmental programs; and changes external to the policy “subsystem,” such as change in socioeconomic conditions, governing coalitions, or constitutional structure. Several hypotheses regarding policy change are specified in this approach.

First, the paper summarizes the policy networks literature and the criticisms levelled against it, particularly the need to develop causal linkages between networks and policy outcomes. Second, the main definitions and hypotheses of the advocacy coalition framework are presented in the context of the network approach. Third, the ACF is applied in a case study of the evolution of Spanish national water policy from 1939 to 2004. Fourth, the advocacy coalition framework is then evaluated in terms of its explanatory usefulness in the Spanish case, and its ability to overcome weaknesses in the policy networks approach more generally.


Policy Networks

As the burgeoning literature on the policy network perspective attests, the concept of network, whether used as a metaphor, a model, a “tool box” (Börzel 1998), or a potential theory, is appealing to many social scientists. This interest stems from dissatisfaction with prevailing models of decision making in the modern state, and from the intuitive understanding that modern policy making is a complex and many-faceted process. The concept of networks—intricate webs of governmental and non-governmental actors at various levels in the system connected by common interests and resource dependencies—thus seems to be able to capture this complexity better than other models (e.g. pluralism, corporatism). McCool summarizes these perceived advantages in his evaluation of the so-called “subsystem concept” (within which he includes issue networks, advocacy coalitions, and policy networks). These advantages include the approach’s ability to move us beyond “a strictly institutional framework that tended to isolate policymakers, when in fact they act in conjunction with a host of policy-influencing entities,” and that these network models “helped move the discipline beyond the stalemated confrontation between elite theory and pluralism” (1998: 553).

Upon entering the policy networks literature, however, one finds a bewildering array of definitions, goals, and approaches. The diversity and complexity of the networks rubric engenders very valid criticism regarding its usefulness and explanatory power (e.g. Dowding 1995, 2001; Kassim 1994). Even summarizing and categorizing the network approach is problematic, with differing perceptions among those who would undertake this task. Network analysis has been classified, for example, as a “typology of interest intermediation” versus a “specific form of government” (Börzel 1998), a “group interaction approach” versus a “personal interaction approach” (Daugbjerg 1998), and, analytically, as both an independent and a dependent variable (Bressers, O’Toole and Richardson 1995).

The most influential and oft-cited works are those that Börzel (1998) categorizes as 1) using the concept as a means of understanding interest intermediation (the British school) and 2) conceptualizing networks as a new form of government/governance in the modern state (the European or German school). These approaches have in common a view of networks as fairly stable patterns of social relationships among interdependent actors (both public and private, at all levels of government), which emerge around policy problems. Policy networks are characterized by both a formal hierarchy of authority as well as formal and informal interactions among these actors (Marin and Mayntz 1991: 18). Resource dependencies among the actors in the network serve as channels for communication, expertise, and other policy resources (Bomberg 1994).

The interest intermediation approach was developed to counter the pluralist versus neo-corporatist typologies of interest intermediation and as an alternative means of categorizing policy making in liberal democracies (e.g. Rhodes 1997: 29-32). Within this rubric, several authors have developed network typologies, which “share a common understanding of policy networks as power dependency relationships between the government and interest groups, in which resources are exchanged” (Börzel 1998: 256).[1]

Policy networks in the governance school are seen as a new form of governance which has emerged as a response to the complexity of the modern state. Similar to Rosenau’s “turbulence” (1992), these analysts conceptualize “governance under pressure” (Jordan and Richardson 1983). As a result of factors such as interdependence and increasingly complex and difficult demands, “modern societies are characterized by societal differentiation, sectoralization and policy growth which lead to political overload” (Börzel 1998: 259) in which “government organizations are no longer the central steering actor in policy activities” (Klijn 1997: 33).

Scholars who see the European Union (EU) as an emerging multi-level system of governance also find the policy networks approach appealing, especially as a counter to state-centric models. The networks concept is increasingly entering discussions on the EU, usually conceptualized as a means of interest intermediation within a multi-level governance framework (see Bomberg 1994, 1998; Peterson 1995a, 1995b). Peterson and O’Toole argue that certain inherent features of EU governance, including the EU’s underlying structure and emphasis on the principle of partnership, suggest the existence of policy networks (2001: 306).

One of the most pressing concerns for all the network literature is establishing better explanations, particularly making causal links between network types or characteristics and policy outcomes. Marsh, for example, argues that “…most authors…see policy networks as having a significant influence on policy outcomes, although they are seldom explicit about the causal mechanisms involved” (1998: 7). Peterson and O’Toole (2001) call for more research that would conceptualize policy networks as both an independent and a dependent variable, that is, investigating causal factors associated with network development and change and in turn, how networks (or network change) affect policy outcomes.

There are some efforts along these suggested lines. Bressers and O’Toole (1998), for example, develop a simple typology of policy networks along the dimensions of “interconnectedness” and “cohesion,” and suggest a model that would correlate network type with government selection of policy instruments. Daugbjerg (1998), in a more complex analysis, provides a revision of the Marsh and Rhodes network continuum and attempts to relate these network types to choice of policy instruments in agri-environmental policy making. Marsh and Smith develop a “dialectical model” in which they explore the relationships between network structure and political agency of the actors within it, between the network and the context in which it operates, and between the network and policy outcome (2000). None of these efforts, however, has succeeded in systematically developing hypotheses regarding the presumed causal links between network type and policy outcome that could be applied more generally to a broad range of cases. Moreover, the kind of empirical analysis required to test such hypotheses, which usually involves extensive fieldwork conducted to map networks and determine actor interactions and preferences, is, as Dowding argues, “time-consuming and expensive” (2001: 89).

An approach which may help us to address some of the shortcomings of network analysis is the advocacy coalition framework. While not falling explicitly into either the interest intermediation or governance camp and having its own theoretical and methodological concerns separate from the policy networks approach,[2] the advocacy coalition framework presents a network conceptualization within which testable hypotheses regarding policy change over time have been developed (see Sabatier 1987, 1988; Sabatier and Jenkins-Smith, eds., 1993).  


The Advocacy Coalition Framework

While the advocacy coalition literature employs the term “subsystem” as opposed to “network,” the meaning of, and factors involved in, the formation of subsystems are the same as those set forth in network analysis. According to Sabatier, “the complexity of modern society, the expansion of governmental functions, and the technical nature of most policy problems create enormous pressures for specialization” (1993: 23). Therefore political elites that are concerned with a specific policy area or problem will form “relatively autonomous subsystems” (Ibid). Other literatures describing such subsystems, for example iron triangles, are deficient in that they are limited to a single level of government and to interest groups, administrative agencies, and legislative committees. A policy subsystem should be considered, then, “the set of actors who are involved in dealing with a policy problem such as air pollution control…” including not only those entities considered by the more narrow approaches but also actors such as “journalists, analysts, researchers and others who play important roles in the generation, dissemination and evaluation of policy ideas as well as actors at other levels of government who play important roles in policy formulation and implementation” (Sabatier 1993: 24). Going beyond the network approach, latent actors and constituencies (not only the currently identifiable network participants) must be considered part of the subsystem.

Given the large and diverse set of private and public actors to be found in each subsystem, Sabatier addresses the problem experienced by network analysts in doing empirical research by “aggregating” actors in subsystems by “advocacy coalitions,” which are “…people from a variety of positions (elected and agency officials, interest group leaders, researchers), who share a particular belief system—i.e. a set of basic values, causal assumptions, and problem perceptions—and who show a nontrivial degree of coordinated activity over time” (1987: 660), thus providing a more manageable focus of analysis.

Within this framework, policy change over time is seen as a function of: 1) the interaction of competing advocacy coalitions within a policy system; 2) conditions/changes external to the subsystem that provide opportunities and obstacles to the competing coalitions (further categorized as relatively stable system parameters such as social structure and constitutional rules and dynamic system events such as changes in socioeconomic conditions or system-wide governing coalitions) (Jenkins-Smith and Sabatier 1993).

The advocacy coalition approach focuses particularly on the belief systems of such coalitions as a major factor affecting the coalitions themselves as well as policy change:

A belief system guides coalition members concerning the problems that should receive the highest priority, the causal factors that need to be examined most closely, and the governmental institutions most likely to be favorably disposed to the coalition’s point of view. The coalition then seeks to alter the behavior of governmental institutions in order to achieve its policy goals over time (Jenkins-Smith and Sabatier 1993: 41).


The framework also distinguishes between “core” and “secondary” elements of belief systems as well as public policies. Coalitions are organized around the “fundamental normative and ontological axioms,” of the former, and are, like these beliefs, therefore assumed to be relatively stable over periods of a decade or more. Secondary aspects of belief systems include “…instrumental decisions and information searches necessary to implement the policy core in the specific policy area” and are moderately easy to change (Sabatier 1993: 30-32).

The hypotheses set forth in the advocacy coalition approach are the following:

  • Hypothesis 1: On major controversies within a policy subsystem when core beliefs are in dispute, the lineup of allies and opponents tends to be rather stable over periods of a decade or so.
  • Hypothesis 2: Actors within an advocacy coalition will show substantial consensus on issues pertaining to the policy core, although less so on secondary aspects.
  • Hypothesis 3: An actor (or coalition) will give up secondary aspects of a belief system before acknowledging weaknesses in the policy core.
  • Hypothesis 4:  The core (basic attributes) of a governmental program is unlikely to be significantly revised as long as the subsystem advocacy coalition that instituted the program remains in power.
  • Hypothesis 5: The core (basic attributes) of a governmental action program is unlikely to be changed in the absence of significant perturbations external to the subsystem, that is, changes in socioeconomic conditions, system-wide governing coalitions, or policy outputs from other subsystems (Sabatier 1993: 27-35).
  • Hypothesis 6: Policy-oriented learning across belief systems is most likely when there is an intermediate level of informed conflict between the two. In such a situation, it is likely that: 1) each coalition has the technical resources to engage in such a debate; and 2) the conflict be between secondary aspects of one belief system and core elements of the other or, alternatively, between important secondary aspects of the two belief systems.
  • Hypothesis 7: Problems for which accepted quantitative data and theory exist are more conducive to policy-oriented learning than those in which data and theory are generally qualitative, quite subjective, or altogether lacking.
  • Hypothesis 8: Problems involving natural systems are more conducive to policy-oriented learning than those involving purely social or political systems because in the former many of the critical variables are not themselves active strategists and controlled experimentation is more feasible.
  • Hypothesis 9: Policy-oriented learning across belief systems is most likely when there exists a forum that is: 1) Prestigious enough to force professionals from different coalitions to participate; and 2) Dominated by professional norms (Jenkins-Smith and Sabatier 1993: 50-54).


Case Study: The Evolution of Spanish Water Policy: 1939-2004

The case of water policy in Spain is a particularly appropriate test of the advocacy coalition framework for several reasons. First, the analysis of Spanish water policy in the context of democratization since the mid-1970s allows us to consider the factors put forth in both the networks literature in general and the advocacy coalition framework in particular for the emergence of networks/subsystems in modern democratic societies. Second, prior studies on Spanish water policy point to a focus on ideas and beliefs such as the regeneracionista (“regenerationist”) concept of conquering nature through grand water projects in the name of modernization (e.g. Swyngedouw 1999, del Moral and Sauri 1999). Third, research in water policy more generally has found that water policy networks are subject to “both endogenous and exogenous change, reflecting shifting conditions (…these include hydrogeological, financial and legal circumstances), values, and participation in the various stages of the policy process” (Bressers, O’Toole, Jr. and Richardson 1995). Fourth, the case allows us to consider the influence of the supranational European Union. Finally, the evolution of water policy in Spain adds to the comparative empirical data already presented using the advocacy coalition framework, such as Munro’s case study of water policy and politics in California (1995).

Spain’s water management and distribution system is one of the largest and most complex in the world and the country has a very long history of human intervention in the “chaotic” patterns of precipitation characteristic of the Iberian Peninsula. These patterns include vast differences in the amount of rainfall in different areas (particularly between the relatively wet Atlantic north and northwest and the arid Mediterranean south), highly variable amounts of rain from year to year and the concentration of rainfall in relatively few days at a time (Maestu 2003: 5-6). In response to these challenges, “Spaniards have been attempting to increase the availability of water for at least the last 2000 years, primarily through irrigation systems that enable them to cultivate dry but otherwise suitable lands” (del Moral and Sauri 1998: 15).

Modern efforts to harness water as a resource for development came to fruition under the Franco regime, when the state became a “master socioenvironmental engineer” (Swyngedouw 1999: 456), increasing the country’s hydraulic capacity by 1000 hm3 per decade.[3] This trend has resulted in the construction of over 1200 dams (Spain has the fourth largest number of dams after the US, China, and India) that regulate around 40 percent of the country’s total renewable water resources per year (Bakker 2002: 775).

This so-called “hydraulic paradigm,” the policy implemented by Franco and successive governments to “develop water resources so that everyone everywhere will have the water they want when they want it” at little or no cost to users (del Moral and Sauri 1999: 12), increasingly has been debated after Spain’s transition to democracy and in the wake of a severe drought in central and southern Spain in the period 1991-1995. While the 1985 Water Law and two subsequent attempts (in 1993 and 2001) to develop a National Hydrological Plan (Plan Hidrológico Nacional or PHN) largely continued this paradigm, the fact that the 1993 plan was defeated and the 2001 plan is in the process of being significantly revised, and that new strategies representing a significant departure from the traditional paradigm have been implemented since the mid-1990s, appear to indicate that policy change is underway.

This paper uses the evolution of the Spanish water policy making process to analyze the factors that contribute to and/or hinder policy change through the lens of the advocacy coalition framework.


Evolution of Policy

While the present case focuses primarily on the period since the transition to democracy, it is impossible to fully understand this period without a longer historical perspective, dating back to the regeneracionista movement in the late 19th century. The Franco period was then characterized by an adherence to this vision of harnessing the country’s water resources through large hydraulic infrastructure projects.


Regeneracionismo: the “social mission” of irrigation

The “regenerationist” movement in Spain was a call for modernization and a reaction against what its proponents—among them literary figures, intellectuals, politicians, and scientists—saw as a decline of the Spanish social, cultural, and political system, marked particularly by the loss of Cuba, Puerto Rico, and the Philippines to the United States in 1898. This movement sought to displace the traditional power structure of caciquismo, embodied in the large land-owing oligarchy, with a strategy that would open Spain to international trade and mobilize its natural resources. A primary component of this strategy was the manipulation or even “creation” of nature, the main focus of which became harnessing the country’s erratic hydraulic resources. As Swyngedouw states, “Spain’s ‘geographical problem’ became the axis around which the sociocultural and economic malaise was explained, and where the course of action resided” (1999: 452). Large-scale state-initiated and financed irrigation systems became this movement’s goal, as “the only way to overcome this difficult environmental, social and economic situation” (del Moral and Sauri 1999: 14).

The advocacy coalition that emerged in expounding this general view was large and diverse, including “reformist socialists, populists, industrialists, and enlightened agricultural elites” (Swyngedouw 1999: 452) and was supported in its need for scientific expertise by hydraulic engineers, represented primarily by the Corps of Engineers, founded in 1799 with primary responsibility for public works in Spain (Ibid.: 457-458). The opposition of conservative, traditional power elites, along with “lack of financial investment, the difficulties of conducting complex agricultural reforms, and the uprising of the Civil War in 1936” (Costejà et al. 2002: 16) largely prevented this vision from being implemented until the Franco era. Several reforms during the early 20th century did show the influence of the regeneracionistas, however. 

The National Hydraulic Works Plan of 1902 (the “Gasset Plan”), for example, had the goal of irrigating around 1.5 million hectares of land through the construction of 205 hydraulic infrastructure projects, though only 150,000 hectares had been irrigated by 1925 (del Moral and Sauri 1999: 15). The first River Basin Authorities (RBAs) were created in the Ebro and Segura River Basins in 1929 and 1931 and given a semi-autonomous role in water resource planning (Maestu 2003: 9). The “Pardo” National Hydraulic Works Plan of 1933 commissioned detailed scientific studies of the various aspects of Spain’s water basins (e.g. climate and agricultural potential) and included many of the infrastructure proposals that were subsequently implemented in the second half of the 20th century (Maestu 2003: 10). This limited progress was spearheaded by the engineers, who came to lead the regenerationist movement “through a combination of their legitimization as the holders of scientific knowledge and their privileged position as a politically elite corps within the state apparatus” (Swyngedouw 1999: 460).


The Franco Era: state as “master socioenvironmental engineer”

The Franco regime embraced the regenerationist strategy of large hydraulic infrastructure projects, but with a somewhat different rationale than that of the modernizers of the 19th century. Providing water to irrigate Spain’s arid southern regions containing a large number of agricultural laborers was part of the “internal colonization” plan designed to improve the economic and social conditions in the poorest areas of Spain, and thereby avoid social unrest (Maestu 2003: 22). The “massive physical mobilization of water” by the state was justified by the “torrentiality of the Mediterranean climate,” (Bakker 2002: 775) and was a primary means of extending the dictatorship’s power and also legitimizing it (del Moral 1999: 32).

Franco drew on the proponents of the regenerationist ideas, such as the Corps of Engineers, for support of his policies. He also centralized the process and limited dissent. The regime built on the existing, quasi-independent River Basin Authorities, for example, but transformed them into instruments of the central state (Maestu 2003: 22). Particularly after the 1959 Stabilization Plan which ended the regime’s experiment with autarky, the resulting development (including increased industrialization and tourism) augmented demand for water across a variety of sectors in addition to agriculture. This only reinforced the agreement among major actors in the regime that promoting and subsidizing large hydraulic infrastructure projects was a necessity in Spain’s continued modernization. The result was a cohesive and stable “water policy community” (Pérez Díaz and Mezo 1998: 4) including public officials in the central government, power companies, and irrigation associations (del Moral and Sauri 1999: 32).


1985 Water Law

The ten years that followed Franco’s death in 1975 saw momentous changes in Spain that certainly would be characterized as “significant perturbations external to the subsystem,” hypothesized by Sabatier to have affects on advocacy coalition beliefs and thereby policy outcomes (1993: 34). The transition to a liberal democracy, decentralization of power to 17 regional governments (Comunidades Autónomas or Autonomous Communities) and local authorities, and preparations to join the European Community (EC) all had important repercussions for Spanish water policy. The “activation of public opinion” (Pérez Díaz and Mezo 1998: 5) resulted in the emergence of groups previously absent from the political discourse, notably environmental groups and academic experts with perspectives that questioned the traditional hydraulic paradigm. Decentralization included the transfer of water administration competencies to regional and local authorities, thereby adding an important territorial dimension, and point of conflict, to water policy. EC environmental legislation as well as economic pressures associated with accession provided important external constraints on policy options.

The 1978 Constitution not only began the process of devolving authority to regional and local levels of government, but also mandated promotion of “the rational use of natural resources,” including protection and restoration of the environment. Given the new power configuration of the democratic regime, the fact that existing water legislation contradicted some of these new constitutional goals, and in the presence of increasing demand for and over-use of water, the reform of water legislation was given priority by the Socialist government (PSOE) after coming to power in 1982 (Costejà, et al. 2002: 17). The primary purpose of the 1985 Water Law was therefore to modernize the legal framework for water policy. It required that water policy cover both groundwater and surface resources within river basins (consolidating the River Basin Authorities as a basis of water management), declared groundwater to be part of the public domain, linked both water quality and environmental protection to the water policy process, promoted participation of water users and affected groups in the decision-making processes, and incorporated EC water quality directives (Law 29/1985). The law also required the Spanish parliament to carry out national hydrological planning, and recognized the authority of the new Autonomous Communities: the plans would be carried out by the regional government where river basins are contained within a single region and by the national government where river basins cross regions.  

Despite these modifications, the basic strategy of harnessing water resources and increasing irrigation was maintained, although the rationale presented by the PSOE government changed somewhat from that of the Franco era, emphasizing state-led water infrastructure as part of a larger welfare policy with the aim of “spatially balanced growth” and greater equity regarding “the access to fundamental natural resources” (Maestu 2003: 24).


Proposed 1993 Plan Hidrológico Nacional

The 1985 Water Law required that hydrological planning be done at the level of individual river basins (carried out at the regional or national level, as specified above) as well as the country as a whole. Moreover, these individual river basin plans had to conform to guidelines established in the national water plan (del Moral and Sauri 1999: 33). Pursuant to the 1985 legislation, the PSOE presented its first draft National Hydrological Plan (Plan Hidrológico Nacional or PHN) for consideration in 1993. After intense public debate, numerous drafts, and despite the severe drought in the south of the country which pointed up the weaknesses in the system and the need for Spain to get its hydraulic house in order, the plan was rejected by the Spanish parliament in 1995.

The PSOE government’s justification of the PHN invoked development and modernization imperatives while focusing on inter-regional solidarity and the “social aims” of achieving equity through increased access to water resources in the arid south. At the center of the plan were large-scale water transfers from the north and west to the south and east, the building of 150 new dams, and a large state investment plan designed to increase Spain’s regulated water resources by over 6000 hm3 a year. Thus, “[i]n its supply-led approach, and in its assumption that the state would continue heavily to underwrite the costs of hydraulic development, the proposed PHN stands in direct continuity with Spain’s 20th-century paradigm of hydraulic development” (Bakker 2002: 776).

A variety of factors internal and external to the water policy subsystem converged in the defeat of the 1993 PHN. As Pérez Díaz and Mezo argue, during the 1980s and 1990s the number of actors involved in the subsystem greatly increased, due to interrelated transformations in the institutional framework of Spanish political life (external factors): 1) the transition to a liberal democracy; 2) the consequent development of the logic of party competition; 3) the activation of public opinion; 4) the decentralization of the Spanish state and empowerment of regional and local governmental authorities. Because of these transformations, “a structure of incentives and opportunities has been created for the appearance of new actors and for modification in the guidelines of conduct of the old” (1998: 6; my translation).

The power of the policy community/advocacy coalition in favor of the traditional hydraulic paradigm was eroded by the reorganization of the state bureaucracy, particularly the weakening influence of the Dirección General de Obras Hidraúlicas (DGOH), the engineer-dominated body within the Ministry of Public Works that had a monopoly on the policy process prior to democratization (Pérez Díaz and Mezo 1998: 6).[4] With this reorganization also came the marginalization of the group of specialists, again dominated by engineers, upon which DGOH relied for expertise, the Centro de Estudios Hidrográficos (CEH). Pérez Díaz and Mezo point out here that most of the technical studies for the 1993 draft PHN were carried out by outside consultants, a significant departure from prior practice (Ibid.). The decentralization process, involving the transfer of authority for aspects of water and environmental policy to the regional level, further weakened the DGOH monopoly on water policy. It also had the effect of “territorializing” the policy area, making it a focal point for the evolving power struggle between the national and regional governments and at times between the regional governments themselves, with water-surplus regions questioning the legitimacy and fairness of transfers and water-deficit regions echoing the central government’s call for national solidarity and development.[5]

The increase in public participation associated with the democratization process increased public debate and brought about the emergence of public and private actors concerned with both the environmental and economic impacts of the PHN, including conservationists, consumer groups, trade unions, and economists. Led by academics and practitioners, “strategic alliances” between free-market economists and environmentalists, in particular, questioned the supply-led engineering approach of the traditional paradigm and the assumption that the economic benefits of increased hydraulic regulation justify the expense (and that the expense must be supported by the public sector), criticized inefficient water use by especially the agricultural sector, and presented demand-led alternatives (Bakker 2002: 776).

Economic concerns within the government were also significant, as indicated by the fact that the Ministerio de Hacienda (Treasury) called into question several of the cost estimates presented in the PHN (claiming that they underestimated the likely true cost) and proposed the creation of a new set of taxes on water demand and other financial mechanisms designed to downsize the public sector (Pérez Díaz and Mezo 1998: 9, 14; Bakker: 777). This position was strongly influenced by the external factor of the economic effects of meeting the convergence criteria for Economic and Monetary Union (EMU), reforms in the Common Agricultural Policy (CAP) and likely decrease in structural funds to Spain in the lead-up to Eastern European enlargement (del Moral and Sauri 1999: 34).

Other external factors constraining the actors in the subsystem were the transposition of European environmental law, particularly the directive mandating Environmental Impact Assessments on major public works (Directive 97/11/EC), and the fact that the proposed PHN had serious implications for transborder water agreements with Portugal, resulting in declarations by the Portuguese prime minister on a visit to its Iberian neighbor in 1996 that the water issue constituted the “principle difference” in policy between the two countries (Pérez Díaz and Mezo 1998: 18).

The political weakness of the PSOE government at what would be the end of its tenure in government also figured into the defeat of the legislation in 1995. In the Congreso, a motion initiated by the opposition conservative Popular Party (PP) required that the PHN could only be approved together with the National Irrigation Plan, which, according to the government itself, was supposed to be elaborated simultaneously by the Ministry of Agriculture (the PSOE accepted this requirement). In the Senado, a motion by the Aragon Regional Party (PAR), seconded by the PP and other opposition groups against the wishes of the government added the stipulation that the PHN could not be approved until the 14 administrative river basin planning units presented their own plans (Pérez Díaz and Mezo 1998: 9, 15).


1999 Modification of the 1985 Water Law

The PP had criticized the PSOE’s inability to overcome the PHN impasse, and after their electoral victory in 1996 promised, through the newly-created Ministry of the Environment (MIMAM), to achieve approval of a national plan. The main vehicle for this approval was to be “enhanced dialogue,” supported by the Libro Blanco de Agua (White Paper), which was produced and disseminated in 1998 (Giansante et al. 2002: 525-26). Moreover, the more conservative, market-based approach of the party led to the application of these principles, particularly privatization, to water policy. This change in philosophy (termed mercantilización or “marketization” by Bakker) was supported by the PP’s articulation of “state failure” and inefficiencies which could be remedied through an aggressive plan of privatization in all economic sectors with the goal of liberalizing the Spanish economy and improving competitiveness. The PP also cited requirements of joining the Economic and Monetary Union (EMU) to justify these policies (Bakker 2002: 782)[6] and promised that water policy would be more sensitive to the regional and environmental characteristics of the resource (Ministerio de Medio Ambiente 1998).

The PP government introduced two initiatives in keeping with the marketization philosophy. The first was the creation of public-private partnerships (Sociedades Estatales) in water infrastructure development. This initiative was made possible by a 1996 reform to the legal structure governing public corporations (Law 13/1996) which introduced new structures of financing for a variety of public works, and involved private financing of the construction and management of water projects chosen by the government (Bakker 2002: 778). The 1999 reform of the 1985 Water Law then created these public corporations in seven Spanish river basins (Maestu 2003: 27).  The second was the introduction of water markets and water banks in the 1999 reform.

The creation of the Sociedades Estatales, as Bakker argues, is not a significant departure from the traditional vision of state-controlled hydraulic policy as an instrument of socioeconomic policy (2002: 778). This “mixed model” is characterized by government financing of less profitable hydraulic works (with support from EU structural funds), and private funding of profitable investments, primarily in urban water distribution (Maestu 2003: 27). The creation of water markets and water banks, however, is a significant innovation which may signal a shift away from more traditional policies.

Water markets allow for a temporary ceding of water use rights among entities and individuals that already hold water rights licenses from the state. These markets “maintain the public water domain institution, but they also establish the possibility of water transfers by means of the temporary surrender of rights from one right-holder to another within a set of limitations”, including that use rights may only be ceded to other rights-holders of equal or higher priority in a river basin’s plan, that non-consumptive licensees (e.g. hydroelectric power operators) may not transfer water rights, and that the state retains the power to reject transfer proposals (Giansante 2002: 531). Water markets are designed to introduce flexibility into the extant system, the weaknesses of which were demonstrated in the drought of the early 1990s.[7] The concept of water banks also is based on the experience with such banks in California, and “are designed to be ‘centers of exchange of water rights’ in cases of drought or other exceptional circumstances,” with the state acting as an intermediary between those seeking the temporary transfer of rights (Giansante 2002: 530).

This new approach of the PP government certainly reflects the party’s ideological stance as a proponent of the free market and privatization, and also their efforts to overcome some very real economic and technical problems apparent in Spain’s water regime, aggravated by the drought and continued increases in water demand. The market-based approach is “an expedient solution” intended to gradually “eliminate the state subsidy of the hydraulic network, to allocate water to (economically) highest value uses, and to incorporate full costs to society and the environment in water charges” (Bakker 2002: 780). The government received support on these policies from not only neoliberal economists but also environmental and other groups that see such moves as a partial solution to the environmental problems created by the traditional state hydraulic projects.


2001 Plan Hidrológico Nacional

The Libro Blanco de Agua produced by the Ministry of the Environment in 1998 was, as initially promised by the PP, “widely consensual” (Costejà 2002: 18). It is a comprehensive document detailing the problems facing the Spanish hydrological system and the reasons behind them, and takes into account all relevant aspects of the issue, including climatic, political, economic, environmental, and international. It also offers recommendations for dealing with what it terms the “crisis of the traditional model,” including the development of alternative sources of supply (e.g. desalination), greater efficiency in use, managing agricultural demand rather than continuing to attempt an increase in supply, reducing state subsidies in water (thereby increasing costs to users), introducing “scarcity indicators” and allowing for private participation in infrastructure financing and flexibility of the permit system (the latter two were the basis for the 1999 reforms) (Ministerio del Medio Ambiente 1998).[8] Despite its support for some marketization, however, it “also raises objections to the indiscriminate use of market criteria” (del Moral and Sauri 1999: 35).[9]

The Ministry of Environment presented a new draft of the PHN in September 2000, which appeared to reject much of the consensus developed in the Libro Blanco and return to state-controlled, supply-led solutions. The new draft reduced the proposed volume of transfers contained in the 1993 proposal by about half (Giansante 2002: 527).[10] Despite this reduction, the plan still envisaged the construction of 120 reservoirs and more than 1,000 kilometers of canals and pipelines. Even though the plan maintains the marketization efforts already carried out and introduces some measures for the protection of natural resources, it very much returns to a command and control system planned and operated by the state. Indeed, the government’s stance on water markets had become “more openly equivocal” at this point, with some PP officials arguing that markets “could not…be deployed to solve the structural imbalance in water resources across Spain” (Bakker 2002: 784). The focus of both support and opposition certainly mobilized around the transfer of water, which would have constituted the most expensive infrastructure project in Europe.

The proposed PHN was voted into law in the PP-dominated legislature in 2001 (Law 10/2001), with little of the public debate promised in 1995. The government suppressed 80 critical scientific reports and ignored protests by opponents (The Financial Times, 13 August 2002). Supporters of the plan comprise many public and private actors within the water-scarce regions, including the regional governments, irrigation farmers, and the tourism industry, as well as many elements of the traditional policy community, e.g. road, canal, and port engineers’ corps, farmers’ organizations and unions, trade unions and some local councils represented by the umbrella organization Plataforma por el Guadalquivir, public works officials in the central government, and construction and hydroelectric companies. 

An increasingly vocal group of opponents emerged to counter the PHN, including national and European environmental groups such as Ecologistas en Acción and the World Wide Fund for Nature,[11] consumer groups, members of the business community who support marketization, many economists, the water “donor” regions represented by the umbrella Plataforma por la Defensa del Ebro (whose residents staged huge protests in Spain and Brussels) left-wing political groups, including Izquierda Unida (United Left) and many members of the PSOE, who are against both water transfers and many aspects of mercantilización.[12] The plan also mobilized a significant number of Spain's scientists against it. In an open letter to the European Parliament, 150 university researchers and professors alleged that the purpose of the PHN was “the promotion of large public works” rather than economic or environmental rationality (The Financial Times, 13 August 2002).

While different opponents stressed different aspects of the plan, several key criticisms emerged: 1) the plan “reverted” to an outdated water culture of command and control and did not reflect the consensus developed in the 1998 Libro Blanco; 2) it contradicted parts of the 1999 modifications to the Water Law in that it had not waited for an evaluation of the results of the water markets in terms of efficiency; 3) it had not considered other possibilities such as water desalination, re-use of sewerage water or water conservation; 4) it is based on an overestimated water demand in the Mediterranean area; 5) it fails to take into account recommendations from the 2000 National Irrigation Plan; 6) it contravenes important aspects of EU legislation such as the Water Framework Directive (2000/60/EC) and other directives requiring environmental impact assessments and the protection of habitats (Costejà, et al  2002: 18).

Much of the opposition to the PHN, both in 1993 and in 2001, was led by academics and practitioners. This is shown not only by the letter to the EP discussed above, but also, importantly, in the expert group that has converged around the Congresos Ibéricos sobre Planificación y Gestión de Aguas (Iberian Congress for Planning and Development of Water), a conference forum supported by 70 Spanish and Portuguese universities dedicated to “scientific rigor, independence, and social sensitivity” (see The group has hosted four major academic conferences dealing generally with the issue of water policy, and spawned the Fundación Nueva Cultura de Agua (New Culture of Water Foundation), which promotes sustainable development principles in water planning and use. Both groups engage in research, publication, debate, and promotion of public awareness of water issues.[13]

As a result of the complaints raised by academics and also environmental and regional groups at the European Parliament, the institution’s Petition’s Committee in 2002 asked the Commission to investigate the claims of non-compliance with EU directives (Europe Information Service, 25 June 2002). The Commission ultimately sent a mixed message to Spain regarding the plan, directly criticizing the country for “wasting water” and recommending an incentive structure that would involve higher prices to users (Expansion, 11 September 2001), but then conditionally approving in early 2004 part of the Structural Fund financing requested by the Spanish government for a limited transfer of water in one area (The Economist, 8 January 2004).


The 2004 Victory of the PSOE: The A.G.U.A. Plan

The PSOE was voted into office rather unexpectedly, in national elections held three days after the Madrid terrorist attacks of 11-marzo. The new Prime Minister, José Luís Rodríguez Zapatero, has changed course on many areas of public policy, including the PHN. In a May 2004 meeting with the European Commission environment commissioner, the new Spanish environment minister outlined the PSOE government’s intention to develop a new water policy. This plan would focus on alternatives to diverting water from the Ebro, including desalination, water treatment and re-use, efficient use of resources and renewable energy. The minister stressed that the water transfer option was not completely off the table, but rather would be regarded as “a last resort” (European Report, 5 May 2004). In June, the government stated that it would “abandon” the water transfer plan in favor of a project involving the construction of desalination plants, and modified the 2001 PHN to allow the PSOE’s planned changes (Royal Decree Law 2/2004).  

The government announced a new plan (referred to as A.G.U.A.) in September 2004. Its proposed cost of €3.789 billion (a projected savings of €400 million compared to the PHN) involves the creation of a “Public Water Bank” in each hydraulic basin, the construction of desalination plants from Catalonia to Almeria, which would provide 60 percent of the annual water requirements of the south, and an effort establish a system of user payments that would better reflect the “true cost” of water. According to Prime Minister Zapatero, the new plan is designed to move Spain toward a “new politics of water…which will take into equal consideration the economic, social and environmental value of water, with the objective of guaranteeing its availability and quality, optimizing its use, and restoring associated ecosystems” (Spanish Environment Ministry website,, my translation).

The PSOE promises to involve the public in debate over this new approach, and along those lines the environment ministry developed a new information section on their website devoted to the A.G.U.A. plan, in addition to publicly announcing the posting of the studies opposing the 2001 PHN (which had been largely ignored by the PP) to the ministry’s website in the name of transparency (see AcuaMed, a new Sociedad Estatal, is charged with carrying out the main aims of A.G.U.A. in the 2004-2008 period and is currently in the process of negotiating the implementation of the plan with users and Autonomous Community governments (del Álamo 2004).

Initial reaction from the opponents of the 2001 PHN has been mixed. The environmental group Ecologistas en Acción, for example, called the decision an “historic victory,” but also stated that “Even if it has much less of an impact on the environment, it will help consolidate a model of non-sustainable development on the Mediterranean coast” (Europe Information Service, 24 June 2004).


Table 1: Advocacy Coalitions in Spanish National Water Policy: Actors and Beliefs

Traditional Hydraulic Paradigm – 1939-1975

MAIN ACTORS: Franco and his advisors; key entities in the centralized executive, e.g. Dirección General de Obras Hidraúlicas, in Ministry of Public Works; road and hydraulic (canal and port) engineers (e.g. Corps of Engineers); Centro de Estudios Hidrográficos; farmers and agricultural laborers, especially in the south; construction and hydroelectric industries


NORMATIVE CORE BELIEFS: Nature is made rational through human intervention and control. Control and transformation of the erratic hydraulic patterns in Spain is a necessity, even a patriotic duty, in order to modernize and develop the country economically and socially. Goal of water for everybody, everywhere, at virtually no cost to users.


POLICY CORE BELIEFS: Water shortages, and therefore poverty, lack of development, and social backwardness, occur because the natural hydraulic patterns in Spain are erratic and result in water waste. Therefore, large-scale, state-directed and funded hydraulic infrastructure projects, primarily dams, reservoirs, and irrigation projects involving transfers of water, are to be vigorously pursued, with a special concentration on the poorest areas for purposes of legitimizing the regime. Decision-making should be limited to central administration and those supporting traditional paradigm.


INSTRUMENTAL POLICY BELIEFS: The benefits from water development greatly outweigh the costs. Agriculture (and later tourism industry) are key in the economic development of the country. Very little policy-related concern for environmental issues.


Traditional Hydraulic Paradigm – 1975-2004

MAIN ACTORS: Conservative political parties (UCD then PP); PSOE (during 1982-1996 term in office); public works officials in central government; governments and River Basin Authorities in water-scarce regions; irrigation farmers and their unions; tourism industry in the south; engineers’ corps; Plataforma por el Guadalquivir (umbrella group of farmers’ organizations and unions, trade unions, local councils); construction and hydroelectric industries


NORMATIVE CORE BELIEFS: Nature is made rational through human intervention and control. Continued development and modernization after the dictatorship relies on maintenance and expansion of hydraulic works, including large-scale transfers. National solidarity is a key component of the transfers, which also have the aim of spatially balanced growth and greater equity in the access to water. Continued goal of water for everybody, everywhere, at virtually no cost to users.


POLICY CORE BELIEFS: Water shortages are due to erratic hydraulic patterns and also increase in demand (which is a necessary part of development). To meet these demands, there is a continued focus on large-scale, state-directed and funded hydraulic infrastructure projects dedicated to increasing supply. Decision-making should now be subject to the democratic process and decentralization of authority.


INSTRUMENTAL POLICY BELIEFS: The benefits from water development outweigh the costs, but there is a recognition that these costs are increasing. Agriculture, tourism and related industries are key in the economic development of the country. Water quality and environmental impact may be considered, especially as relates to the 1978 Constitution and EU environmental and water policy.



ACTORS: Left-leaning members of PSOE; Izquierda Unida and other leftist parties; Spanish and European environmental groups; governments in water “donor” regions; some consumer groups; researchers and scientists (e.g. Fundación Nueva Cultural de Agua)


NORMATIVE CORE BELIEFS: Nature should be respected, not controlled or manipulated. Ecological values should be given consideration equal to or greater than economic growth goals. Human actions in the environment should be constrained for the current and future public good. Moral argument that water is a public, not a private, good, as it is essential to life itself. Sustainable development criteria should always be applied.


POLICY CORE BELIEFS: The traditional paradigm, especially large-scale water transfers, results in degradation of natural environments and habitats, and promotes waste rather than conservation. The true costs of water use should be taken into account and particularly large users should pay. All groups and citizens affected by state water policies must be included in the decision-making process.


INSTRUMENTAL POLICY BELIEFS: The costs of water development, both financial and ecological, outweigh the benefits. The agricultural sector in particular wastes water. Efforts should be directed toward conservation, efficient use of water, development of alternatives such as desalination, and higher water costs (although these should not fall unduly on citizens, and the “commodification” of water should be avoided).



ACTORS: many PP politicians; Treasury officials; some consumer groups; sectors of business community (such as waste-water treatment companies) benefiting from public-private arrangements (Sociedades Estatales); economists


NORMATIVE CORE BELIEFS: The market is highly valued as an efficient allocater of all types of goods. The problems faced currently in Spain regarding water demand and supply, high financial costs, and inefficiencies are due to state failure in allocating resources. Free market values and economic liberalization are the key to dealing with these inefficiencies. Water should be considered a “normal” commodity.


POLICY CORE BELIEFS: Mercantilización will result in water conservation, stimulate supply, and help the agricultural sector to modernize and improve efficiency.


INSTRUMENTAL POLICY BELIEFS: The financial costs of supply-led water development outweigh the benefits, and should not be borne by the public sector. The agricultural sector in particular wastes water. Water infrastructure development and delivery should be privatized. Supports initiatives such as Sociedades Estatales, water banks, and water markets, but would like to see more market mechanisms introduced. Full cost recovery is a goal, and all users should pay.


Source: Author’s elaboration based on framework in Sabatier 1993, Table 2.1, p. 31 and Munro 1993, Table 6.1, pp. 116-117.


Advocacy Coalitions and Policy Change

In the advocacy coalition framework (ACF), policy change within a subsystem (dependent variable) is the product of: 1) competing advocacy coalitions attempting to translate important aspects of their belief systems into governmental programs, by increasing policy resources and engaging in policy-oriented learning; 2) events external to the subsystem that affect the resources and constraints of subsystem actors (independent variables) (Sabatier 1993: 34). Key hypotheses related to these two general causal factors (specified above) will be analyzed in the context of the case study.


Policy Change in Spanish Water Policy

In order to evaluate the independent variables specified in the ACF framework, it is first necessary to identify changes that occurred in Spanish water policy during the period of study. We can say that little significant change took place in Spanish water policy from the consolidation of the Franco regime in the late 1930s until at least the early-1990s, marking an extremely long period of policy continuity. While the regeneracionista ideals pre-dated the dictatorship, it was not until after the Civil War that the policy goal of “water for everybody everywhere” provided through large hydraulic infrastructure projects came to fruition. Although water policy was modified at the margins to take into account the democratization process and new national and international concerns with the environment (the 1985 Water Law), the traditional hydraulic paradigm continued to drive policy even after the Socialists came to power in 1986, as indicated in their 1993 PHN.

The period from 1993 until 1999 was characterized not by policy change, but rather by policy impasse, as shown by the 1995 defeat of the PSOE-proposed national water plan. Subsequent to the PP electoral victory in 1996, the 1998 Libro Blanco recommendations and the more concrete 1999 modification of the Water Law may be viewed as a change in the direction of “marketization,” (with the creation of the Sociedades Estatales, water markets, and water banks) but the PP’s 2001 PHN then largely returned to the traditional paradigm. The new PSOE government’s rejection of the hydraulic infrastructure development of the 2001 PHN, and their A.G.U.A. alternative, would seem to entail a significant shift in the direction of the environmentalist preferences of conservation and desalination, while maintaining some marketization goals, particularly regarding user costs (but showing a preference for public, rather than private, water banks). It remains to be seen, if course, how much of the A.G.U.A. initiative will be implemented.


Advocacy Coalitions and Policy-Oriented Learning

We now ask the question of what effects the various advocacy coalitions—and the interactions and policy learning taking place within and between them—have on the identified changes (and non-changes) to Spanish water policy.

The longevity of a water policy based primarily on harnessing resources through large infrastructure projects is largely attributable to the stability and strong normative core belief system of the traditional hydraulic paradigm. The strength of this commitment to the philosophy of human alteration of the environment as necessary for modernization was conceived at an intellectual, and almost spiritual, level during the regenerationist period. It was then solidified in policy terms, and strengthened in intellectual terms, during the long period of dictatorship in which the regime relied on this “state as master socioenvironmental engineer” stance to legitimize its hold on power, and in which dissenting voices were excluded. This case thus supports the following ACF hypotheses: that the lineup of allies in an advocacy coalition tends to be stable over periods of a decade or more (Hypothesis 1); and that the core of a governmental program is unlikely to be revised significantly as long as the advocacy coalition that put it in place remains in power (Hypothesis 4).

The rather drastic exogenous political, structural, economic, and socio-cultural changes that occurred as part of democratization (discussed in the next section) were thus necessary in shaking the stability of the traditional hydraulic paradigm and allowing the development of advocacy coalitions with dissenting views. The traditional paradigm still plays a role but has been weakened particularly since the 1990s, as competing coalitions (the environmentalists and marketizers) have emerged, with the result that water policy has begun to reflect some of the norms and goals of these new coalitions.

The core values that the environmentalists and marketizers have developed are based largely on scientific and economic data that challenge many of the core “engineering” norms of the traditional paradigm. While the normative core beliefs of the new coalitions differ significantly from each other (see Table 1), there is room for compromise, and for allying strategically against the traditional coalition, because some of their policy core and instrumental core beliefs are compatible. The primary areas of convergence involve the beliefs that the costs of water development embodied in the traditional paradigm outweigh the economic benefits, that the agricultural sector wastes water, and that the true costs of water should be considered and users should pay (though there is difference between the coalitions regarding how far the commodification of water should go). So even though the primary core concern of the environmentalists is respect for nature and promotion of sustainable development, they can support some aspects of mercantilización as a means to the end of water conservation. And conversely, even though the normative core beliefs of the marketizers focus on faith in liberal economics as a solution to all water difficulties, they seek the support of environmentalists and have been willing to compromise in promoting partial, and slow, market solutions that do not go “too far” in environmentalist terms. This type of cooperation between the new advocacy coalitions supports the ACF hypotheses that the normative core values are the most difficult to change, and that compromise, change, and learning across belief systems are more likely in the instrumental policy, and policy core beliefs (see Hypotheses 2, 3, and 6).

The marketization-oriented reforms carried out by the PP government in the late 1990s also support these hypotheses.[14] While not giving up the normative core belief in the necessity of large, state-directed hydraulic projects in the name of development, many government officials came to see the introduction of Sociedades Estatales, water markets, and water banks as a way to maintain these overarching goals while dealing with the growing concern about government expenditure on the projects, increase in demand, inefficiencies in the delivery of services, and compliance with EU legislation. The PP thus attempted to “deploy mercantilización as a technical facilitator of the continuation of the traditional hydraulic paradigm” (Bakker 20021: 780), that is, adjusting instrumental policy beliefs while maintaining the normative and policy core. In engaging in this type of “policy learning,” however, the traditional paradigm coalition thus gave more legitimacy to those actors in the marketizer and environmentalist coalitions who wish to move further away from traditional policies, particularly in the area of cost recovery.

The general process of policy learning and the role of scientific research and knowledge in that process (see Hypotheses 6-9) are also supported by the Spanish case. The ACF predicts that a coalition which opposes a proposed action or program will try to mobilize latent actors to join their coalition by actively pointing out the costs of the proposal to all relevant actors. Further, coalitions will accumulate and mobilize technical and scientific evidence in an attempt to change core beliefs that underlie policies with which they disagree. These strategies have been used fairly effectively by both the environmentalists and the marketizers, as they sought to widely publicize the economic and ecological costs of continuing the hydraulic paradigm in the 1993 and 2001 national hydrological plans. Moreover, the accumulation and dissemination of economic data regarding the cost/benefit ratio of large projects and environmental impact data (as actively pursued and disseminated by, for example, the Fundación Nueva Cultura de Agua) has resulted in some policy learning within the traditional coalition as well as support for alternatives on the part of an increasing number of societal and political actors and the general public.

The role of knowledge and science has been a prominent factor in the strengthening of particularly the environmentalist coalition. At the forefront of their efforts have been academics and researchers coming together through professionalized fora such as the Fundación and the Congresos Ibéricos sobre Planificación y Gestion de Aguas, and then using the growing body of evidence on environmental damage and inefficiencies of the traditional paradigm to take an activist role in promoting alternative norms and policy preferences. Arguably their efforts are aided by the nature of the problem, which allows the empirical documentation of the shortcomings of extant policy in terms of water shortage issues and environmental degradation, particularly in the wake of the drought.

An important aspect of this case that is difficult to account for through the advocacy coalition framework is the territorial dimension introduced by decentralization and transfer of authority to the regional governments. In effect, water-scarce regions (encompassing both public and private actors) entered the traditional paradigm coalition in strong support of large-scale transfers, while water-rich regions did battle on the side of anyone who would advocate limiting or stopping these transfers, thus throwing in their lot with the environmentalists and marketizers. The ACF does take into account the purposeful mobilization of latent actors by existing coalition members. However, its focus on belief systems, as opposed to interests, does appear to fall short in explaining the strong territorial conflicts that have significantly affected the water policy subsystem.[15] The regional actors involved have not subscribed to any normative aspect of the competing belief systems; rather, they are acting strictly on the basis of economic interest, that is, are we likely to be hurt or helped by water transfers? They have couched these interests in terms of ideals such as national solidarity or regional patriotism, but their underlying economic interests are what drive their position and actions. This territorial dimension has been very important in understanding particularly the policy impasse in the early 1990s (contributing greatly to the intractability of the issue), but cannot be explained sufficiently by the ACF.


Impact of Changes External to the System

The other crucial variable that affects policy change, according to the ACF, is the role of factors external to the policy subsystem. Hypothesis 5 makes this quite clear: “The core of a governmental action program is unlikely to be changed in the absence of significant perturbations external to the system…” (Sabatier 1993: 34). Therefore, when two cores conflict, each tends to use scientific and technical analysis primarily to buttress its own arguments, resulting in a “dialogue of the deaf” (and thus policy impasse) “until external conditions dramatically alter the power balance within the subsystem (Jenkins-Smith and Sabatier 1993: 48). So while the “enlightenment function” provided by research is a necessary condition for significant policy change, it is not sufficient; exogenous shocks to the subsystem are required as well.

These external factors are divided in the model into two categories: 1) relatively stable parameters, which include basic attributes of the problem area, basic distribution of natural resources, fundamental cultural values and social structure, and basic constitutional/legal structure (rules) and 2) dynamic system events, which include socioeconomic conditions and technology, systemic governing coalitions, and policy decisions and impacts from other subsystems (Sabatier 1993: 18-23).

In Sabatier’s view, the so-called dynamic system events “constitute the principal dynamic elements affecting policy change” since these types of changes alter the opportunities and constraints that confront the actors in the policy subsystem (Ibid.: 22), while the more stable parameters serve to limit generally, over a period of several decades, the range of alternatives available to the actors. Certainly the case of Spanish water policy shows the importance of external factors. Moreover, it affords us an opportunity to examine rather sudden changes to the “relatively stable parameters” in the democratization process after 1975.

The 1978 Constitution drastically altered the entire legal structure of the Spanish system and the fundamental cultural values and social structure also shifted toward democratic norms fairly rapidly. The result of these changes was to greatly increase the number of actors involved in the water policy subsystem and make possible the emergence of the environmentalist and marketizing advocacy coalitions. These groups, along with regional and local actors, were empowered through both the formal authority granted to them in the new democratic system and the opportunity they now had to build up other power resources, such as public support and legitimacy, technical and scientific expertise, and economic resources.

Accession to the European Community in 1986 also had profound impacts on the stable parameters and dynamic system events pertinent to the water policy subsystem. Spanish water policy would now be subject to a growing body of European environmental legislation on a variety of issues such as habitats and environmental impact assessment, and most recently on all aspects of water (the Framework Directive). This supranational legal structure and commitment constrained the actions of the traditional coalition. Evidence that policies resulting from the hydraulic paradigm ran counter to EU goals and legislation bolstered the claims of those groups that were seeking alternatives, and gave them another point of influence above the state, as shown by the mobilization in Brussels of various Spanish groups opposing water transfers and the PHN more generally. This influence was reciprocal, as the Spanish government was subject to criticism of its water policy from European institutions such as the European Parliament and the Commission, and European environmental groups such as the WWF. Moreover, accession requirements and Spain’s later commitment to meet the EMU convergence criteria, and the likelihood of declining structural funds in the face of Eastern European enlargement put constraints on the public spending necessary to continue the traditional reliance on state-funded construction of large infrastructure projects.

The drought of the early 1990s also had the effect of altering the basic attributes of the problem area and basic distribution of natural resources, at least in the southern regions. This natural event resulted in a hardening of core attitudes, however, rather than an opportunity for policy learning. Water shortages were used as evidence by both the hydraulic paradigm coalition and its challengers that their approach was the correct one—the former advocating even greater water transfers from the north and the latter stressing the need for cost recovery, a more flexible system of water rights, and conservation.

The elections of 1996 and 2004 altered the “systemic governing coalitions” in ways that can be predicted by the ACF framework. The PP’s analysis of welfare state failure and the party’s commitment to economic liberalization and concern to rein in state spending resulted in the exploration of new policy responses of the 1998 Libro Blanco and the subsequent mercantilización reforms. The PSOE victory in 2004 has thus far ushered in a commitment to the party’s more left-leaning goals, including environmentalist concerns with conservation, sustainable development, and alternative water sources (e.g. desalination), reflected in the A.G.U.A. plan.

The policy decisions and impacts of other subsystems also have constrained policymaking in the water sector. The agricultural/irrigation sector has been particularly important. While pressures from large irrigators support the traditional paradigm, a growing consensus in the government that the national irrigation and water plans should be harmonized was a partial cause of the 1995 defeat of the PHN. Also, environmental policy formulated and implemented from the supranational to the local level increasingly constrains policy options in the water sector.



Conclusions: Wider Implications for Network Analysis

Strengths of the Model

Aggregating actors into advocacy coalitions as a unit of analysis does appear to achieve the goal of making the empirical data gathering process manageable without sacrificing explanatory ability. Little of value would have been added to the current study by a more detailed mapping of the actors involved.

The model addresses another important criticism of the network approach, that causal relationships—either addressing change in networks or change in policy—are seldom specified or researched sufficiently in most of the networks literature. The ACF conceptualizes policy change as a result of two main variables, the competition of advocacy coalitions attempting to translate their belief systems into policy (by increasing their resources and engaging in policy learning), and exogenous shocks to the subsystem that constrain and enable the actors in these efforts. Importantly, it brings in the role of scientific and technical knowledge, and its use, to explain policy-oriented learning. This is a particularly relevant part of the explanation of policy change in the Spanish case, and in the so-called information age it would appear to be applicable more broadly.

The advocacy coalition framework is able to provide explanations for policy change without the need to construct the typologies typical of the network approach. It is not the characteristics of the network actors and their relationships (such as numbers of participants, stability of membership, insularity versus permeability of the network) that affect policy change. Rather, change is a function of how successful competing advocacy coalitions are in attempting to translate important aspects of their belief systems into governmental programs (by increasing policy resources and engaging in policy-oriented learning), and how changes external to the subsystem constrain or enable them in this endeavor. Given the difficulties involved in constructing typologies, the disagreements regarding what characteristics should be included in them and an overall lack of hypothesized causal linkages between network type and policy change, the ACF would appear to be the more fruitful approach.

The advocacy coalition approach also would seem to provide a more coherent explanation of network change than presented in the policy networks literature. In the latter, policy networks are seen as emerging as the result of “governance under pressure” factors, that is, increasingly complex and difficult demands on the modern state leading to political overload. But we have few efforts to specify why and how this works to create policy networks, or whether variances in these types of factors might affect the type of networks that emerge (which in turn may affect policies/policy change). The ACF seems to get us around this problem by hypothesizing that such factors external to the policy subsystem are a necessary impetus to policy change since they then alter the constraints and opportunities of the advocacy coalitions, who will then take advantage of these changed circumstances to try to mobilize support for policy based on their preferred norms.

Finally, the Spanish case demonstrates that the ACF may be used to consider the effects of the EU without having to explicitly “add” the supranational level to the model. Rather, the focus on advocacy coalitions as containing actors at all levels of governance, coupled with the consideration of external factors, allows the incorporation of the EU into the analysis in a coherent and seamless way.


Weaknesses of the Model

Despite the relative manageability of using advocacy coalitions as the unit of analysis, the model remains far from parsimonious. We might argue that this is necessary in explaining complex modern systems, but we still risk getting bogged down in the details of actor beliefs and preferences and in classifying who belongs in which advocacy coalition (and run the risk that different analysts will classify these in different ways). Moreover, the sheer number of hypotheses and detailed explanations for the mechanisms behind each one calls into question whether these are really testable, or rather are simply an extended description of policy processes as conceptualized by the model.

As indicated above in the evaluation of the Spanish case regarding the territorial dimension to the water transfer conflict, the emphasis on belief systems, as opposed to interests, may leave out an important explanatory variable. The ACF rejects the view that the primary motivation of policy actors is (especially short-term) self-interest. While acknowledging that “belief systems are normally highly correlated with self-interest and the causation is reciprocal,” ACF proponents argue that belief-system models are more inclusive: “Interest models must still identify a set of means and performance indicators necessary for goal attainment; this set of interests and goals, perceived causal relationships, and perceived parameter states constitutes a ‘belief system’” (Sabatier 1993: 28).  But then how do we accommodate actors, such as the water-rich and water poor regions in Spain, who came to play such a crucial role in the policy debate yet whose motivation clearly was economic self-interest as opposed to a belief in the policy norms of any existing advocacy coalition?

Despite these weaknesses, the strengths of the advocacy coalition framework would seem to make it one of the most useful tools available for those who remain convinced that the complexities of policymaking in multi-level systems are best captured by the network approach.



Álamo, Maribel del, 2004, “Entrevista: Adrián Baltanás, Director General de AcuaMed,” (Interview: Adrián Baltanás, General Director of AcuaMed), Ambienta, no 38 (November).

Arrojo Agudo, Pedro, ed., 2001, El Plan Hidrológico Nacional a Debate, Bilbao, Spain: Bakeaz/Fundación Nueva Cultura del Agua.

Bakker, Karen, 2002, “From state to market?: water mercantilización in Spain, Environment and Planning A 34: 767-790.

Bomberg, Elizabeth, 1994, “Policy Networks on the Periphery: EU Environmental Policy and Scotland,” Regional Politics and Policy 4, no. 1 (Spring): 45-61.

——, 1998, “Issue Networks and Environment: Explaining European Union Environmental Policy,” in David Marsh (ed.), Comparing Policy Networks, Buckingham, UK: Open University Press.

Bomberg, Elizabeth and John Peterson, 1998, “European Union Decision Making: The Role of Sub-national Authorities,” Political Studies XLVI: 219-235.

Börzel, Tanja A., 1998, “Organizing Babylon—on the Different Conceptions of Policy Networks,” Public Administration 76 (Summer): 253-273.

Bressers, Hans and Laurence J. O’Toole, 1998, “The Selection of Policy Instruments: a Network-based Perspective,” Journal of Public Policy 18, no. 3: 213-239.

Bressers, Hans, Laurence J. O’Toole, Jr., and Jeremy Richardson, 1995, “Networks as Models of Analysis: Water Policy in Comparative Perspective,” in Bressers, O’Toole, Jr. and Richardson, eds., Networks for Water Policy: A Comparative Perspective, London: Frank Cass.

Costejà, Meritxell, Nuria Font, Anna Rigol, and Joan Subirats, 2002, “The Evolution of the National Water Regime in Spain,” Universitat Autonoma de Barcelona, accessed from EUWARENESS Project (funded by the European Commission, coordinated by the University of Twente in The Netherlands) website:

Daugbjerg, Carsten, 1998, Policy Networks under Pressure: Pollution Control, Policy Reform and the Power of Farmers, Aldershot, UK: Ashgate Publishing Company.

Dowding, Keith, 1995, “Model or Metaphor? A Critical Review of the Policy Network Approach,” Political Studies XLIII, 1995: 136-158.

——, 2001, “There Must be End to Confusion: Policy Networks, Intellectual fatigue, and the Need for Political Science Methods Courses in British Universities,” Political Studies 49: 89-105.

Giansante, C., M. Aguilar, L. Babiano, A. Garrido, A. Gómez, E. Iglesias, W. Lise, L. Moral, and B. Pedregal, 2002, “Institutional Adaptation to Changing Risk of Water Scarcity in the Lower Guadalquivir Basin,” Natural Resources Journal 42 (Summer): 521-563.

Jenkins-Smith, Hank C. and Paul A. Sabatier, 1993, “The Dynamics of Policy-Oriented Learning,” in Sabatier and Jenkins-Smith, eds., Policy Change and Learning: An Advocacy Coalition Approach, Boulder: Westview Press.

Jordan, Grant and Klaus Schubert, 1992, “A Preliminary Ordering of Policy Network Labeling,” European Journal of Political Research, 21: 7-28.

Kassim, Hussein, 1994, “Policy Networks, Networks, and European Policy Making: A Skeptical View,” West European Politics 17, no. 4 (October): 15-27.

Kickert, Walter J.M., Erik-Hans Klijn, and Joop F.M. Koppenjan, eds. 1997, Managing Complex Networks: Strategies for the Public Sector, London: Sage.

Klijn, Erik-Hans, 1997, “Policy Networks: An Overview,” in Kickert, Klijn, and Koppenjan, eds., Managing Complex Networks: Strategies for the Public Sector, London: Sage.

Marin, Bernd and Renate Mayntz, eds., 1991, Policy Networks: Empirical Evidence and Theoretical Considerations, Boulder, CO: Westview.

Marsh, David, 1998, “The Development of the Policy Network Approach,” in Marsh, ed., Comparing Policy Networks, Buckingham: Open University Press.

Marsh, David and R.A.W. Rhodes, 1992 “Policy Communities and Issue Networks: Beyond Typology,” in Marsh and Rhodes, eds., Policy Networks in British Government, Oxford: Clarendon Press.

Marsh, David and Martin Smith, 2000 “Understanding Policy Networks: Toward a Dialectical Approach,” Policy Studies 48, no. 1: 4-21.

Maestu, Josefina, ed., 2003, “Public Participation in River Basin Management in Spain: Reflecting changes in external and self-created context,” Workpackage 4 of the HarmoniCOP Project, University of Alcala de Henares, Spain.

McCool, Daniel, 1998, “The Subsystem Family of Concepts: A Critique and a Proposal,” Political Research Quarterly 51, no. 2 (June): 551-570.

Ministerio de Medio Ambiente, 1998, Libro Blanco de Agua (White Paper on Water), Madrid: Ministerio de Medio Ambiente. Also available at:

——, 2003, Water in Spain (revised and translated version of the 1998 Libro Blanco de Agua, Madrid: Ministerio de Medio Ambiente.

Moral, Leandro del and David Sauri, 1999, “Changing Course,” Environment 41, no. 6 (July/August): 12-36.

Munro, John F., 1993, “California Water Politics: Explaining Policy Change in a Cognitively Polarized Subsystem,” in Sabatier and Jenkins-Smith, eds., Policy Change and Learning: An Advocacy Coalition Approach, Boulder: Westview Press.

Pérez Díaz, Victor and Josu Mezo, 1998, “Política del Agua en España: Argumentos, Conflictos y Estilos de Deliberación” (Water Policy in Spain: Arguments, Conflicts, and Styles of Deliberation), Primer Congreso Ibérico sobre Gestión y Planificación del Agua, Fundación Nueva Cultura de Agua, Zaragoza, accessed:

Peterson, John, 1995a, “EU Research Policy: The Politics of Expertise,” in Rhodes and Mazey, eds., The State of the European Union, Volume 3: Building a European Polity? Boulder, CO: Lynne Rienner Publishers.

——, John, 1995b, “Policy Networks and European Union Policy Making: A Reply to Kassim,” West European Politics 18, no. 2 (April): 389-407.

Peterson, John and Laurence J. O’Toole, Jr., 2001, “Federal Governance in the United States and the European Union: A Policy Network Perspective,” in Nicolaidis and Howse, eds, The Federal Vision: Legitimacy and Levels of Governance in the United States and the European Union, Oxford; New York: Oxford University Press.

Rhodes, R.A.W., 1997, Understanding Governance: Policy Networks, Governance, Reflexivity and Accountability, Buckingham: Open University Press.

Sabatier, Paul A., 1987, “Knowledge, Policy-Oriented Learning, and Policy Change,” Knowledge: Creation, Diffusion, Utilization 8 (June): 649-692.

——, 1988, “An Advocacy Coalition Framework of Policy Change and the Role of Policy-Oriented Learning Therein,” Policy Sciences 21: 129-168.

——, 1991, “Toward Better Theories of the Policy Process,” PS: Political Science and Politics 24, no. 2 (June): 147-156.

——, 1993, “Policy Change over a Decade or More,” in Sabatier and Jenkins-Smith, eds., Policy Change and Learning: An Advocacy Coalition Approach, Boulder: Westview Press.

Sabatier, Paul A. and Hank C. Jenkins-Smith, 1993a, Policy Change and Learning: An Advocacy Coalition Approach, Boulder: Westview Press.

Swyngedouw, Eric, 1999, “Modernity and Hybridity: Nature, Regeneracionismo, and the Production of the Spanish Waterscape, 1890-1930,” Annals of the Association of American Geographers 89, no. 3: 443-465.

van Bueren, Ellen M., Erik-Hans Klijn, and Joop F. M. Koppenjan, 2003, “Dealing with Wicked Problems in Networks: Analyzing an Environmental Debate from a Network Perspective,” Journal of Public Administration Research and Theory 13, no. 2 (April): 193-213.

Wilkes, Stephen and Maurice Wright, eds., 1987, Comparative Government-Industry Relations: Western Europe the United States and Japan, Oxford: Clarendon Press.

[1] Network typologies are developed along dimensions such as number of participants, stability of membership, insularity versus permeability, and relative strength or weakness of resource dependencies. See, for example, Rhodes 1997: 36-45; Wilkes and Wright 1987; Jordan and Schubert 1992.

[2] For example, this approach was developed largely to combat what its proponents call a “stages heuristic to public policy” derived from the work of David Easton and Harold Lasswell. See Jenkins-Smith and Sabatier, 1993.

[3] 1 cubic hectometer (hm3) = 1 million cubic meters (m3); 810.71 acre-feet (ac-ft)

[4] The DGOH was steadily weakened through a series of reforms of the Public Works, Transport and Environment Ministries beginning in 1977.

[5] The so-called “water wars” have pitted donor and recipient regions, and their PSOE regional presidents, against each other (and the donors against the national government as well) in a bitter battle over the logic and necessity of proposed transfers.

[6] The PSOE had engaged in privatization measures as a matter of economic necessity, but due to their overall philosophy, had proceeded with caution, on a case-by-case basis, i.e. privatización vergonzante—“shameful” or “bashful” privatization (Montoro 1993, cited in Bakker 2002: 781).

[7] For example, cases emerged in which farmers continued to irrigate their crops with little restrictions while urban users experienced restricted use and even cessation of water service (Bakker 2002: 779).

[8] The 1998 Libro Blanco was updated and revised in 2003. The revised version does not depart significantly from the 1998 goals and recommendations, however.

[9] This caution was not enough for the Spanish Economic and Social Council, a consultative organ of the government, which thought that the document went too far in suggesting the “commodification” of water resources (del Moral and Sauri 1999: 35).

[10] The new plan concentrated on transfers from the Ebro river basin to the Mediterranean basins, eliminating the transfers from the Tagus and Douro basins contained in the 1993 proposal (and which are shared with Portugal).

[11] See, for example, “Draining the life out of the land,” The Ecologist 32, no. 1, Fall 2002; “Good Water Management—Experiences and Alternatives to the SNHP” (World Wide Fund for Nature-Europe:

[12] These groups, along with some associations of small and medium-sized farmers and environmental groups, have criticized water markets on the basis of the potential negative socio-economic effects (e.g. concentration of resources in sectors of highest productivity, thereby hurting smaller, poorer farmers) and also the “moral argument” that water should remain a public rather than private good, as it is essential to life itself. They argue, instead, for reduction of waste through water losses and leaks, and increased flexibility in the agricultural sector. The PSOE, while in opposition, stated a preference for public water banks (Bakker 2002: 784).

[13] See also Arrojo Agudo, ed. 2001.

[14] In particular, proponents of the ACF framework argue that policy oriented learning is “an ongoing process of search and adaptation motivated by the desire to realize core policy beliefs…” and that actors will try to respond in a manner that is consistent with their normative and policy core beliefs when confronted by challenges to their belief system (Jenkins-Smith and Sabatier 1993: 44-45). While change may occur, it is very difficult (even in the presence of exogenous shocks) for actors to alter their core beliefs, not least because these are largely normative and therefore not readily susceptible to empirical challenge.

[15] Indeed, the framework “explicitly rejects the view that actors are primarily motivated by their short-term self-interest and thus that ‘coalitions of convenience’ of highly varying composition will dominate policy making over time” (Sabatier 1993: 27).