1. Introduction

The provision of urban water services is an essential service that provides a basic good critical to life. Since the early 1990s, important changes in the structure of water service provision mean that the politics of access to water, especially in relation to water service delivery, are increasingly regulatorypolitics. Whether water services are delivered by public or private entities, they increasingly operate at arms length from traditional representative politics and under the supervision of some kind of regulatory body. Yet the politics of access to water are also a passionate flashpoint for popular social and cultural concerns, often expressed in the language of human rights or through the action of direct protest.

The participation of civil society is often presented as a ‘solution’ to both the technocracy of regulatory politics and the unruliness of populist direct action. This article explores both the direct and indirect roles of civil society in the regulation of water services, drawing on examples from research conducted in 2003-4 in Argentina, Bolivia, Chile, France, New Zealand, and South Africa.¹The findings emerge from events that occurred during the decade between 1995-2005. The research focused on jurisdictions that have involved the private sector was involved in the delivery of water services, even if only partially. The research design focused in detail on one or two disputes in each case study country in order to draw out the possibilities for routinising disruption.

On the one hand, civil society can assist in shaping regulatory frameworks by helping to routinize procedures and institutional interactions – often through the introduction of legislative, regulatory and custom-based practices that provide a stable framework for actors with widely diverging values to interact. On the other hand, civil society organizations are often important for the organization and expression of more ‘ unruly’ politics, channelling disruption and direct protest to test the boundaries of taken-for-granted choices about regulatory frameworks. There is an important dialectic between routinization and unruliness, and the article explores the role of civil society not only in shaping regulatory frameworks for the delivery of water services but also in challenging those frameworks through activism. Case studies that foreground disruption and direct provide ‘limit cases’ to discussions of governance that too often prioritise routinisation, and bring the important dimension of agency to the fore (Morgan 2007a).

The relationship between routinization and unruliness can be understood better by categorizing the gaps between regulatory and citizen space that emerge in the case studies according to key contrasts that underlie them. One contrast, especially salient to the perspective of civil society actors, is between strategies and mechanisms that seek to build political agency on the one hand, and those that aim to embed responsible consumer behaviour on the other hand. Another contrast, especially salient to the perspective of the regulatory actors that civil society seeks to influence, is between ‘transactional’ and ‘ political’ regulation. Although these contrasts are related to each other, the gulf between citizen concerns and the approach of regulators is often wide, and not necessarily simply as a consequence of familiar problems such as information asymmetry. Rather, there may be profound differences in perspectives about what counts as an important problem, as well as the solution to such problems. In particular, when regulation is perceived as building a framework for stable transactions between government and service provider, the role of the eventual user, consumer or citizen becomes displaced. As a result, their concerns are either marginalized, or made the province of a discretionary relationship between service provider and ‘end-user’, a relationship that is beyond the bounds of regulatory control.

The article shows how comparative analysis of case study narratives can contribute to a process of building bridges between regulatory and citizen space. ‘Bridges between regulatory and citizen space’ are mechanisms, strategies or institutional design choices that respond to disruption and protest in ways that routinise their impact, while still remaining responsive to the concerns expressed by the protestors. The findings illustrate that civil society becomes involved in the regulation of water service delivery in three principal ways: via traditional representative politics, via technocratic or expertise-based decision-making, and via directly participatory politics. The introduction of a regulatory agency into the policy mix does not have a systematic impact on the type of civil society involvement. Local-level governance of water services is correlated with civil society use of the judicial system, and developed countries offer more space to civil society in traditional representative politics. The article concludes with some tentative generalizations about when – and why – the involvement of civil society can bridge the gap between regulatory and citizen space in the provision of essential services. Where civil society involvement successfully shapes the regulatory framework for the production of water services, there is more chance of a political agency model being built, whereas when civil society seeks to influence the terms of consumption, it generates a responsible consumer model. To build successful bridges, tensions between the two models must be balanced, but numerous institutional vehicles can achieve this, as will be illustrated by the case studies, and by Tables A and B in the conclusion.

2. Civil Society Involvement in the Regulation of Water Service Delivery

The core of this article summarises findings from a cross-national research project focusing primarily on six case studies which were selected to vary along a number of different dimensions that explore a cross-section of possible governance contexts. They all involve one or more of what were, at the time of the research, the three largest multinational water companies. They include both developing countries and OECD countries (Argentina, Bolivia, Chile, France, New Zealand, South Africa), and a full range of different legal structures (three concessions, one management contracts, one privatisation, one public-private partnership). What follows are six brief and inevitably partial and compressed narratives that highlight different aspects of civil society involvement in the regulation of water service delivery in the six case studies. I take civil society in this context to refer to associations, organizations, or other collectivities (including spontaneous mobilizations of individual citizens) who play, or attempt to play, an explicit role in shaping the terms of water service delivery, but are not charged with either delivering the service or with directly regulating the framework for its delivery.

The accumulated narratives that follow in this section aim to provide a basis upon which I attempt to systematize and generalize findings in the conclusion. However, the dispute-centred nature of the research means that the patterns identified in the case studies are not necessarily representative of the way things are done in the case study country as a whole. Nonetheless, all the interviews and documentary analyses were conducted with a careful ear for the specificity of the selected disputes, and in two cases (Chile and South Africa), those disputes originally selected were considerably de-emphasised. As a result, the Chilean case study focused instead on general patterns of regulatory governance in relation to water services, and the South African case study took a preliminary look at ongoing disputes in Durban and Johannesburg. With these caveats in mind, the table on this and the following page summarises the range of material originally selected (and in some cases later restructured) for research.

2.1 Chile

The case study of Chile stands out by virtue of the relative absence of major disputes over private sector involvement in the delivery of water services. This does not mean such involvement has gone uncontested, which is far from the case, but during the period of research the contestation rarely spilled over into unruly dimensions on the street nor was taken to court. Nor did civil disobedience in the sense of non-payment campaigns prevail. At the same time, civil society participation in more routine, technocratic processes (in particular the five-yearly tariff reviews conducted by the regulator) was extremely minimal: the process was highly technical and quantitative and although consumer input was technically permitted, in practice it was neither encouraged nor demanded.

This reflects a general trend of relatively low levels of civil society participation in policymaking and implementation in social and economic affairs in Chile more broadly, particularly in contrast with political and human rights issues. In 2001, then President Lagos put in place an extensive programme for strengthening civil society, but its activities quite strikingly did not focus on issues of infrastructure or major political economy issues, but rather on social welfare, education, health, and neighbourhood community service issues. Chile also has a central sector-specific regulator for the water sector, but although there was vigorous interaction – both in terms of negotiations and in terms of formal dispute resolution, including judicial suits – between the regulator, the government and water providers, there was strikingly little civil society involvement in these processes, which are markedly technical and complex (the legislation even has complex econometric equations relating to the tariff calculation model written directly into its clauses)

By contrast with this civil society quietism, legislative battles over the terms of the regulatory framework for the delivery of water services were persistent and this tended to be the leverage point for what civil society participation there was. Particularly since full privatization began in 1998, legislative proposals from individual congressional deputies proliferated, with many of them aimed at ameliorating some of the distributive effects of privatizing water. For example, one aimed to clarify whether individual citizens can draw water from wells in response to tariff rises by the public water provider. While many or even most of these proposals failed (as a result of executive dominance of the legislative process), one particularly important battle succeeded. In 2004, essential services of water and electricity were included in the consumer protection legislation and made subject to a mild form of class actions (Engel 2005). This change occurred after years of ardent lobbying, partly by state agencies and in particular the SERNAC (Servicio Nacional del Consumidor), but also by civil society groups supporting that process. This included, oddly enough, the Central Workers’ Union which developed a Consumer Division in hopes of having a stronger voice in a policy environment it found overwhelmingly hostile to traditional labour issues.

Another, less formal and in that respect rarer, instance of civil society involvement related to the situation in the provincial city of Concepcion, where an unofficial referendum was organized to protest the sale of Concepcion’s water utility to the then British-German company RWE-Thames. This appears to have had no discernible substantive impact either on the specific situation or on the subsequent trajectory of the structure of water service delivery, although it contributed to collective organising strategies that continue to be used, for example in relation to the consumer protection reform mentioned above.

Two further observations are important: first, that Chile does have a system of household subsidies that cushion the impact of rising tariffs for poor households, though it is administered as a top-down bureaucratic initiative with little if any civil society involvement. Secondly, it should be noted that the situation in the rural areas was quite different: here, user committees were involved in managing the delivery of water services and collecting operational costs from users, as well as liaising with the regional sanitary service companies for maintenance, repair and expansion of service installations, activities funded by the central state through the Ministry of Public Works’ Hydraulic Works Directorate (Alegria and Celedon, n.d). In this area, further developments were later proposed by a leading environmental organization in Santiago for developing a certification system whereby user committees in collaboration with providers and state officials would certify rural providers as environmentally beneficial (and possibly as socially inclusive), blending autonomous control over those services with central oversight of standards.³Interest in certification has expanded significantly during and since the period of research, standards relevant to the delivery of water service developed by the International Standards Organisation (Morgan 2006d) and the Water Stewardship Initiative⁴are becoming increasingly salient to the global debate.

2.2 France

France is an interesting case study as the French water market is dominated by two of the world’s largest multi-national providers of water services, previously known as Suez (now Ondeo) and Vivendi (now Veolia). Yet the institutional structure of water service delivery in France is highly decentralised and until the 1990s, almost completely unregulated by any central norms. Possibly as a consequence of this striking mix (a prominent global presence dominated by a few companies on the one hand, and decentralized locally embedded national delivery on the other hand ), the involvement of civil society in French water services policymaking splits broadly down reformist and radical lines (Morgan 2004a).

There is considerable civil society participation of a reformist kind in France’s more globalrole in the issue of access to water. This occurs at the grass-roots level: for example, NGOs such as Programme Solidarite Eau⁵who provide technical assistance with service delivery on the ground in developing countries). It also occurs at the level of policy formulation, where ‘civil society’ tends to include major businesses working collaboratively on policy issues, most visibly in the French-based World Water Council in Marseille,⁶but also in more national institutions such as the Institute of Delegated Management,⁷and to some extent in connection with the activities of the International Office of Water⁸and the French Academy of Water.⁹

Civil society participation in the regulatory dimensions of water service delivery at the national and local level in France is on the whole fragmented and small-scale, but much more conflictual than in relation to global-level policymaking, and tends to involve a different network of actors. While civil society does contest the involvement of the private sector in the provision of water in France, this has been a relatively recent development, not yet mirrored by formal mechanisms for the participation of civil society. This is striking given that social activism in civil society with respect to issues of public service, essential services and the social entitlements of the republican state is highly salient in France (Ancelovici 2002) and that an important strand of the French ethos of public service involves continuing state responsibility for universal access and equality (Clarke 2000).But this ethos seems paradoxically absent in relation to the delivery of water services, which lack even extensive regulatory structures, much less significant central state involvement. Rather, it has been left to local municipal governments to either provide the services directly themselves or to negotiate contractual arrangements with private providers. Since the early 1990s, however, some basic legal regulation by statute (the Sapin law¹⁰)of the contractual process has been put in place at legislative level. This regulatory framework primarily aims to impose transparency requirements on the process of negotiating or renegotiating contracts between municipal governments and private companies, a process which before the 1990s was shrouded in secrecy and scandalised by the occasional exposure of abuse.

One particularly high-profile scandal regarding the contract for the provision of water services in the regional city of Grenoble ended in a corruption trial that some argue (Laime 2003) was instrumental in creating the political space for passing the Sapin law.. Whether or not this is the case, two very different forms of ‘civil society’ involvement can be noted since the media exposure of dubious contractual negotiations in cities like Grenoble. The first is the formation of local grass-roots organizations (for example Eau Secours in Grenoble) who, in the absence of formal mechanisms for participating in local government decisions,¹¹operate through a combination of ties with minority local politicians (e.g. the Greens) and the exploitation of carefully documented lawsuits.. Meanwhile, a second and very different form of (arguably) civil society involvement shapes events at the structural level of making key decisions about whether or not to delegate the operation of water service delivery. That is the mediating role of non-governmental consultants (Public Service 2000), a national organisation with regional chapters, originally sponsored by the professional mayors’ associations, who assist, for a fee on a non-profit basis, local governments of small towns in monitoring the legal requirements of contracting for water services required by the Sapin law. This is an interesting structure which may be transferable but, like the certification initiative in Chile, is a highly professional and technocratic arena for civil society participation.

2.3 Argentina

Argentina, as a federal country, assigns responsibility for water service delivery to provincial level. During the 1990s, many of its provinces established regulatory agencies to oversee that delivery, and a substantial number delegated the delivery of water services by concession to the private sector.¹²Although the regulatory entity created to monitor the concession in the capital city of Buenos Aires formally provided for consumer participation, no other regulatory structure in the Argentinian water sector did so, and the dispute that was the subject of the research, the Tucuman concession, is therefore more typical in Argentina. More specifically, the typical situation is one where there is little or no formal provision for civil society involvement either in the process of making macro-structural decisions about the organisation of water service delivery, or in the day-to-day operations of regulatory oversight. This is despite the existence of a federal constitutional obligation (Article 42 of the 1994 Constitution) that requires user association participation in regulatory frameworks for the supply of public utilities.

While that constitutional obligation is rarely elaborated in provincial legislative and regulatory structures however, it played an important role in more indirect methods of civil society participation. For example, in Tucuman, the delegation of water service delivery to the international private sector in 1995 was the direct catalyst for the formation of non-governmental consumer associations who contested the involvement of the private sector and the terms on which water was available subsequent to that involvement. As part of that contestation, widespread payment boycotts were instituted, along with more directly visible forms of protest such as marches and demonstrations. While these may appear to be disruptive of everyday regulatory procedures rather than contributing to their smooth operation, the ultimate intention was also constructive – arguably, the disruption created organizational and political space for making alternative proposals. Thus the two major consumer associations in Tucuman both wrote numerous proposals for amending the terms of water service delivery to make them more acceptable to the constituencies they came from (they did not claim to represent them in any formal sense). And these proposals, while not taken up directly in the regulatory arena, received official imprimaturs from time in other institutional arenae. In particular, congressional deputies took up some of the proposals, and the Ombudsman played a vital role on two fronts – first, in lodging court cases that challenged key aspects of the terms of delivery currently on offer, and secondly, in partially legalizing the behaviour of payment boycotters – or at least temporarily protecting them from being pursued legally by the providers (Morgan 2006b).

The forms of civil society involvement that were catalysed initially by private sector involvement persisted into the era of public service operation that returned after the Tucuman concession failed. Civil society promoted many legislative proposals, particularly relating to tariff structures, disconnection and the recovery of unpaid bills, both before and after the return of water services into public hands. Civil society also continues to pursue court cases such as a complex suite of litigation supported by the Ombudsman on the question of whether non-payment was justified in light of deficiencies in service delivery, or was a form of civil disobedience that would attract legal reprisals (Morgan 2006b).

Notably, the civil society activity documented in the Tucuman case was predominantly at the local level over day to day operational terms, although it was initially catalysed by the structural decision to bring in the private sector. Despite this linkage, civil society involvement on structural issues, including the international litigation that followed upon Vivendi’s withdrawal from Tucuman, has been fairly minimal. This is in stark contrast with Bolivia.

2.4 Bolivia

Bolivia, and specifically the Cochabamba concession, has become an iconic case for activists around access to water, and a reviled example for the broader water policy community. While the Cochabamba concession undoubtedly had significant design faults that catalysed its failure, the case still illustrates some fascinating dynamics of civil society involvement in the regulatory framework for water services. One of its most striking features is the gap between formal provisions for the involvement of civil society and the actual practice. Unusually for observed disparities between rules and practice, both formal and actual provision illustrate substantial civil society involvement, but on rather different terms. Formally, there was extensive provision for civil society involvement in the routine operation of local government decision-making by virtue of the 1994 Law of Popular Participation, which created Oversight Committees at local level composed of representatives from local grass-roots groups with formal legal power comparable to that of a mayor, including the ability to suspend disbursements from central to local government if it judges that funds are being misused (Faguet 2003). These parallel, corporatist forms of social representation for civil society have been compared to an upper house of parliament (Faguet 2003), but in practice they played little role in the trajectory of water service delivery in Cochabamba.

This is because at the same time as the Oversight Committees were created, the government also created a regulator, the Superintendent of Basic Sanitation, who reported to the central government Ministry of Housing and Basic Services. In 1997, this regulator was given powers that centralised the control and supervision of service provision that had until then been up to municipal discretion (local mayors had traditionally headed the Boards of Directors of public water companies). When the regulator moved under these new powers to establish a private concession contract for water services in Cochabamba, the Interinstitutional Commission structure established to negotiate with the primary bidder did include, albeit unofficially, the President of the Civic Committee, but was otherwise entirely composed of government officials.¹³

Although the Civic Committee was a non-governmental association purportedly representing the views of civil society, the fact was that extensive civil society involvement already existed in managing water service delivery and had done so for decades - from women-run local water committees to organizations of peasant irrigators. These forms of civil society did not accept the formally sanctioned role of associations such as the Civic Committee, but their insistence on participation in any case extended far beyond the more unruly street protests that are usually the focus of comment on this case. In particular, detailed alternative proposals were frequently presented to government officials and constantly revised. For example in 1999, a coalition of indigenous groups presented a counter-proposal to the water resources law that was then being debated,¹⁴and in October 2004, an Irrigation Law (Ley numero 2878, de Promoción y Apoyo al Sector Riego)largely drafted from the bottom up with the participation of the organizations of small farmers and with the support of the Federation of Small Farmers and Community Systems of Potable Water of Cochabamba (FEDECOR), was passed. This law, amongst other things, created a new water authority called the National Irrigation Service (Servicio Nacional de Riego – SENAR) that included participation and oversight by the small farmers and major small farmer organizations. This agency was to take over the decisions on water resources from the regulator (Superintendent for Basic Sanitation), whose primary jurisdiction was that of water service delivery.

In short, just as the reform trajectory in Bolivia saw formalized politics gradually gain control over previously autonomous grass-roots processes, so in reverse did those organizations participate in formalized politics even where no seat at the table existed for them. Notably they used both direct action on the streets and, albeit initially unsanctioned, the traditional methods of formalized politics (lobbying with detailed legislative proposals) and combined them in such a way that the former eventually carved out a space for the official acceptance of the latter.

Two other forms of civil society involvement in the Bolivian case should be noted, at very different levels. One is the presence of civil society representatives on the board of the public water company structure that replaced the failed private concession. This had some interesting early effects in building links with grass-roots water committees in the southern zones of the city that lack access to piped water. However, civil society involvement tended to depend heavily on specific personalities and after an initial spurt of energy, tailed off in the latter part of the research period (Terhorst 2003). The second form was the extensive networking with international civil society that has made Cochabamba such an iconic case. This only developed since the conflict actually spilled over into violence in 1999, and while its contribution to the detailed elaboration of day-to-day management issues was smaller than initially hoped for,¹⁵it played an important role in focusing attention on the international arbitration case that resulted from the termination of the concession. The case, which claimed US$25 million in damages and US$25 million in lost profits from the Bolivian government, was settled in 2006 for the symbolic sum of 2 Bolivianos in the wake of years of international civil society campaigning, including a petition to the World Bank endorsed by citizen groups from 43 countries. While it is not possible to say whether or not the settlement was a direct result of the civil society advocacy on the issue, Bolivia remains a striking instance of civil society combining direct action, lobbying and traditional representative politics, and action in expert professional fora.

2.5 New Zealand

The public policy framework for water service delivery in New Zealand during the 1990s was shaped, at least in part but importantly so, by vigorous and sometimes unruly consumer activism that appealed in part to the notion of human rights, additionally invoking a range of other arguments and strategies that drew on the special status of water as an essential good or service (Morgan 2006a). The role of consumer activism was institutionally effective only in combination with two other crucial features of the institutional environment: small political parties, and autonomous (and fragmented) local government. Arguably, an unorthodox use of the courts by the activists also played an important role (Morgan 2007c). The combined effect of these institutions and actors was to secure legislative change that constrained both the scope and extent of private sector participation in water services, as well as prohibited the disconnection of citizens who do not pay their bills.¹⁶

Pressure to expand the role of the private sector in delivering water came from both local and foreign business interests. But water in New Zealand is a responsibility of regional and local government which has strong autonomy, both traditionally and in terms of financial capacity, and each council has had the discretion to make separate governance arrangements for the delivery of water. Although two major reforms of the 1974 Local Government Act during the first half of the 1990s increased pressure on local governments to demonstrate value for money in all their activities and to justify explicitly decisions not to use the private sector in providing services to local communities, few extended this logic to water services. Indeed, in the political climate of the mid-1990s, public determination to establish water as a kind of ‘final frontier’ of marketization policies helped contribute to the formation of the Alliance party, a radical left party that won electoral success at regionalgovernment level in the capital city, Auckland, on a platform of anti-privatization. In the subsequent political battle between local and central government that followed, the Alliance Party succeeded in passing legislation that explicitly prohibited the privatization of the bulk water supplier for Auckland (Watercare).¹⁷

There was substantial citizen opposition to taking a commercial or private route to gaining such economies of scale. The Auckland Water Pressure Group (WPG) formed in 1996 with a fluctuating membership of as many as 2000 people, mainly from lower working class families, to fight both private sector participation and commercial restructuring of public water services.¹⁸In 1997, Papakura District Council – the smallest of the six – signed a 30 year concession contract with the foreign private sector, and in the same year Auckland City Council corporatized its water services, creating Metrowater, a local authority trading enterprise fully owned by the local council but that must conform to company law governance structures.

These developments catalyzed a trajectory of consumer activism that not only derailed other councils’ plans to corporatize but also resulted, by 2002, in legislative constraints on privatization at the level of local government, supplementing the outcomes secured at regional level by the Alliance Party. In essence, the WPG’s small-scale but rambunctious social activism created a political space for the minority members of government upon whom the Labour depended to pass legislation, allowing them to insert – into the otherwise open-ended and facilitative 2002 Local Government Act – restrictions on the involvement of the private sector in water service delivery.¹⁹While this was an intentional outcome of the activists’ activities, a further unintentional but equally important legislative amendment was also secured: water disconnections were prohibited and water restrictions allowed only where it would not create unsanitary conditions.²⁰

The WPG challenged the commodification of water through an unorthodox mix of civil disobedience, legal strategies, savvy use of media and political lobbying. Their approach mixed a practical reliance upon (some would say distortion of) administrative law and consumer rights litigationwith a rhetorical public emphasis on human rights, environmental and social justice issues. The interesting feature of this mix was that in relation to the ‘hard law’ regulatory framework, the activists used their status as consumers of a commercial service to press their objectives, but combined this with ‘soft’ activities that vigorously rejected this commercial framework. Moreover while their strategies of direct action exposed them to some degree of public censure for undercutting norms of responsible consumer behaviour, their capacity to air their arguments in court (even while losing) legitimized their cause, and at the same time amplified the publicity they received regarding their civil disobedience. In this way, the activists reshaped policy terms through disruptive challenge and yet simultaneously routinized conflicts over water service delivery.

2.6 South Africa

Civil society involvement in South Africa water service delivery policy during the period of research had three principal dimensions: a ‘high politics’ dimension of constitutional litigation; an ‘ordinary politics’ strategy of advocacy and feedback through NGO lobbies at policy levels and ‘ water fora’ at operational levels; and a grass-roots politics of direct action.

South Africa has to date explicitly avoided creating an independent regulatory agency to oversee the delivery of water services, maintaining direct political control at least over the framework for provision, even if not always over service delivery itself. That political framework has, however, over time, been supplemented and arguably partially displaced by a more transactionally-focused approach. This was reflected during the research period in the establishment of a Municipal Infrastructure Investment Unit, greater legal controls given to the Treasury,²¹dilution of an initial statutory preference in favour of public provision,²²and by the negotiation of arrangements from time to time with the private sector that sidestepped a legislative mandate to involve affected groups in the political policymaking process over water.²³This trend towards a more transactional approach largely tracked a steady increase (that has since tailed off) in the extent of private sector participation in delivering water services, at least as measured by population coverage.²⁴

The overall result of this can be summed up as a hard law framework that was increasingly transactional and relatively neutral to the identity of the provider, combined with ‘soft law’ (policy-based, non-statutory) measures to legitimate this approach. The latter did not, for the most part, reinscribe opportunities for political participation and influence into the regulatory framework, but rather ameliorated its harshest side-effects. It was the impact of social activism at the local level, including important activism around consumption practices, that performed such a reinscription. Organized labour, partly through its role in the governing ANC coalition, partly through lobbying, but also through strike and protest action, played the most important role in bringing the legislative and regulatory framework into its uneasy mix of contradictory signals, as well as umbrella coalitions of NGOs such as the South African Water Caucus. But in the townships and peri-urban areas, citizen and consumer resistance to the move towards greater cost-recovery and marketization in the delivery of water had equally important implications for the ongoing viability of the framework changes. In these areas, there had long been severe problems of mass non-payment for services, the result of collective political action taken by township residents in protest against apartheid. With apartheid ended, cost recovery principles applied to previously badly underserviced areas, even in diluted form, raised tariffs very significantly from the low base flat rate that was charged (but not paid) under apartheid. Township residents boycotted payment, and in relation to water, employed a wide mixture of strategies to disrupt the policies of the government, including marches, protests, payment boycotts, illegal reconnections, political education and test case constitutional litigation.

While the use of different strategies is shaped strongly by both local political dynamics and the social characteristics of different activist groups (Morgan 2007b), there was also a broad and important contrast between strategies that sought to build political agency and strategies that aimed at embedding responsible consumer behaviour. Tensions between these strategies created barriers to productive interactions that might have built bridges between regulatory and citizen space.

During the period of research, strategies focused on political agency had more mass support, and had very different goals from those associated with ‘responsible consumer behaviour’. Building political agency, for example, was a goal of mass mobilisation strategies, both cooperative peaceful ones and adversarial violent ones that co-existed in a pattern one participant called ‘popcorn politics’.²⁵This activism aspired to establish vaguely specified alternatives to capitalism and rejected altogether models premised on private sector participation. The aim was to harness the ‘politics of sheer refusal’²⁶into a more pro-active, mundane, sustainable political education that could create a sense of collective identity for those excluded not just from basic provision in water, but also in health, education and shelter.²⁷Whether or not these activists aspired to the level of representation of a political party, they were motivated by pragmatic service delivery issues (service standards, the cost of water) primarily because of broader goals to change the structural agenda of neoliberalism and privatisation.

The starkest contrast with this species of civil society involvement was provided by ‘tri-sector partnership initiatives’, for example ‘consumer education’ programmes that were run by Durban Metro Water Services in partnership with Vivendi (Veolia) Water. Seeking to build social and political consensus in support of current reforms, the partnerships employed local young people to work with their neighbours on paying bills, managing debt schedules, water conservation techniques, the proper operation of sanitation systems and the like. The structural questions that were the concern of the more disruptive activists were part of the taken-for-granted background for this work. While this work was initially envisaged as technical and professional rather than political, a striking change in job description for those liaising between citizens and the two partners in service provision occurred mid-programme. ‘Customer Service Agents’ were in the second phase of the project renamed ‘Community Development Officers’. This reflected the early inefficacy of a technical, problem-solving approach, and the realization by the partners that securing consensus in the community required a less instrumental approach and one more sensitive to developing collective political agency.

It should be noted that there was no necessary link between individualised modes of assistance and consumer identity on the one hand, or collective strategies and political agency on the other. For example, citizens also mobilised in what some call ‘ten-rand marches’, converging in large crowds on the offices of the water service provider with ten rand notes in their hands to symbolise what they are willing to pay per month for water. This was a strategy that melded political agency and consumer responsibility while still making a strong appeal for addressing the structural forces that lead to a situation where ten rand is the maximum affordable amount an individual can pay.Legal actions also blended collective and individual strategies in interesting hybrids. Activists’ use of legal strategies in South Africa to try and ameliorate the impact of market-based water service delivery on poor consumers initially met with only limited success. Neither of the two cases brought during the period of research succeeded in eliminating disconnection for non-payment of bills as an option for the state or the private water provider. The second²⁸ provided some temporary relief by mandating a range of procedural protections for the consumer before disconnection. Insofar as the first case appeared to withhold a remedy because the plaintiff had illegally reconnected to the network,²⁹the cumulative effect of the two cases was to provide important but purely procedural protection to citizens who paid what they could afford, and who refrained from civil disobedience in their broader demands to the political decision-makers. The litigation had no effect on the principal issue that divided the stakeholders in the broader structural conflict: the justice or appropriateness of a cost-recovery approach to the delivery of water services.

During the period of research on which this article is based,³⁰attempts by activists to pursue in court the guarantees of basic human rights encoded in the formal legislative framework did not reinforce the identity of active political agent, but rather supported a model of ‘soft consumerism’ which softened the harshest effects of market strategies but left broader questions of collective representation and responsiveness to be determined by political institutions that increasingly adopted transactional model of service delivery. This left activists who sought to assert a collective political identity with no ‘everyday’ forum to do so, only the use oppositional and unruly forms of participation that were increasingly repressed by the formal political system.

3. Comparative Analysis

I conclude by summarizing the trends depicted in narrative form in the preceding sections from two connected angles: civil society involvement and regulatory framework design. Linking these angles makes it possible to envisage effective bridges between regulatory and citizen space. First, I will draw some general conclusions about the patterns of civil society involvement in the different case studies. The analysis evaluates the role of factors such as the existence of a regulatory agency, the relative centralization of water services policy, and the status of developing or developed country, and discerns two principal models of civil society involvement: political agency and responsible consumer. Secondly, I will point to a different, albeit related, design choice in relation to a regulatory framework for the provision of water services, which can be either political or transactional.

3.1 Civil society involvement

In Table A, ten different forms of civil society involvement in the regulation of water service delivery across the six country case studies are summarised. The box indicating the presence of such a form is checked only if it is salient (present to a significant degree and reasonably influential on outcomes) in a particular country. A bracketed cross indicates either a weaker degree of involvement via the relevant form, but one that is still salient or likely to become important; or a situation where there is a fairly substantial amount of such activity but it has very little influence on outcomes.

Table A

As shown by Table A, France and New Zealand show the greatest density of strategies for involving civil society. This may be a consequence of their relatively high levels of socio-economic and industrial development and consequent fiscal and institutional capacity to support more avenues for such involvement. But notably, neither France nor New Zealand possesses a sector-specific regulatory agency (the case studies are evenly divided on this important feature: those marked R on the table do have regulatory agencies for the water sector). Is it possible, then, that that the presence of a regulatory agency is associated with a narrower range of strategies for involving civil society in the development of policy frameworks, since at least in the cases of Bolivia and Chile, this is a clear contrast to New Zealand and France?³¹

Further consideration of the evidence indicates that the presence or absence of a regulatory agency does not have a clear impact on civil society involvement. The narrowing of strategies observable in the Bolivian and Chilean cases took substantially different forms. In Bolivia, it concentrated on direct action and mass mobilisation strategies, generating collective contributions to policymaking that only crystallize into formally recognized processes after sustained pressure from the direct action. By contrast, in Chile the narrowing of strategies focused on more formal traditional politics or relatively invisible negotiations within the regulatory agency. These differences echo broader national histories and institutional contexts and are thus unlikely to be caused by water-sector-specific factors. Moreover, when we look at the regulator-freecase studies, there are also differences: South Africa exhibited a narrower range of civil society strategies than New Zealand and France. The differences show that the presence or absence of a regulatory agency has no necessaryimplication for involving civil society in water service delivery policy.

If the presence or absence of a regulator has little definitive effect on patterns of civil society involvement, can any other structured effects be discerned? Further analysis is assisted by clustering the strategies detailed in Table A. Modes 1-4 are, at least when employed by civil society, loosely connected with traditional representative politics; modes 5-8 with more technocratic or expertise-based decision-making, centred on dispute resolution; and modes 9 and 10 with more directly participatory forms of politics. Traditional representative politics, technocratic or expertise-based decision-making, and directly participatory politics represent different institutional avenues for involving civil society in the regulation, broadly understood, of water service delivery. They are all essentially procedural mechanisms. From this perspective, some broad trends emerge:

1. First, cases where a central regulatory agency is involved do not have much less of direct participatory politics than cases without a regulator, implying that a regulator does not necessarily absorb or routinise social conflict over access to water issue. The exception is Chile.
2. Secondly, the developing country examples on the whole have less civil society involvement in traditional representative politics than the developed countries
3. Thirdly, judicial strategies taken by civil society seem to be more prominent in the cases where primary responsibility for water service delivery frameworks is at the local or provincial level (Argentina, South Africa, New Zealand, France) rather than central government level (Bolivia and Chile).

But the full implication of civil society involvement for policy outcomes can only be understood by focusing on the ways in which these procedural possibilities are used to shape substantive policymaking inputs (what and who is being regulated) and outputs (what substantive goals are pursued). The various combinations possible can be organized by reference to two major models of civil society involvement that have different capacities to build bridges between regulatory and citizen space: a political agency model and a responsible consumer model. There is an intuitive link between each model and different kinds of policymaking outputs. Thus for example, substantive goals of basic needs and human rights would link to a political agency model, consumer entitlement goals link to a responsible consumer model, and social justice and poverty reduction goals to one or the other, depending on the level of participation involved in the procedural avenue used by civil society.

But I would argue that the case study narratives explored in the preceding sections of this article show that the more important cleavage is between different kinds of policymaking inputs, and in particular between the regulatory framework for the production of water services, and the terms upon which those services are consumed. Where civil society involvement successfully shapes the regulatory framework for the production of water services, there is more chance of a political agency model being built whatever the substantive goal being pursued, whereas when civil society seeks to influence the terms of consumption, it generates a responsible consumer model, again notwithstanding the explicit goals pursued. Table B, which includes a brief link to examples from the narrative case studies, summarises this argument visually.

Table B

3.2 Regulatory frameworks

Just as patterns of civil society involvement indicate a distinction between political agency strategies and responsible consumer strategies, so too do the details of regulatory frameworks for the provision of water services divide into two key approaches: political and transactional. Transactional frameworks minimize political discretion (for example over tariff-setting processes) and emphasize the allocation of property rights and protection against risk (primarily for those funding infrastructure operation and investment), value for money, affordability and open procurement procedures. In essence, transactional frameworks seek to establish a framework for market-based exchanges that will in the process deliver access to water for individuals. Political frameworks by contrast either consciously modify market-based exchanges, or directly specify redistributive outcomes that could not or would not be achieved in a market setting. In the process they preserve political discretion (especially on key issues such as tariffs) and prioritize consultative and participatory mechanisms for communicating with labour and consumers over the structure of water services.

Political and transactional frameworks provide different degrees of opportunity and responsiveness for the key actors in water policy networks. In particular, transactional frameworks prioritize the relationship between funders, governments and water operators (producers), whereas political frameworks have more space for the voice of civil society as it is broadly defined in this paper. Transactional frameworks can accommodate, and indeed often seek, civil society input. But they tend to insert it into the province of a discretionary relationship between service provider and end-user that is both ‘beyond the bounds’ of regulatory control and highly individualized. For example, high tariffs in a transactional framework are dealt with as individual affordability matters related to the income levels of an individual customer (individual debt negotiation) rather than as a policy issue related to the structural capacity of large groupings within society to bear different levels of fees or cross-subsidise each other (collective bargaining over distributive issues).

Individualisation has implications for ‘feedback’ between transactional and political frameworks of regulation, and it is this feedback that is crucial for resolving tensions between the political agency and responsible consumer models of civil society involvement. This is because where a regulatory framework or legal system is designed to individualise the entitlements and obligations of the actors it controls, it reduces the bases of collective action, and thus narrows the possibilities for political agency to operate. For example, assigning individual title to formerly collectively owned resources, as was on the agenda in the Bolivia water law reforms, fragments the collective owners and pits them in competition against each other (Houtzager and Kurtz 2000). If collective action on the part of groups other than producers and regulators is repressed or restricted in a regulatory system, important linkages between regulatory and citizen space can be compromised.

Take for example, the effect on Tony Prosser’s well-known argument (Prosser 2000) that the regulatory framework introduced by the UK when it privatized essential services, including water, has had unexpected benefits in terms of protecting vulnerable end-users. The core of his argument is that the institutional vehicle of independent regulators can enhance the chances of achieving poverty reduction goals, by enabling them to evolve as an integral part of a stable and predictable baseline environment for investors, as opposed to being perceived as political arbitrariness. Put in the terms of this paper, ‘transactional’ regulation when it directly addresses end-user and especially vulnerable end-user concerns can provide a bridge between regulatory and citizen space without actually becoming ‘political’ regulation, with all its attendant risks from the service provider perspective.

However, when transactional regulation targets vulnerable end-user concerns using responsible consumer strategies of involving civil society, there is a risk that neglecting the cultivation of political agency on the part of the beneficiaries of such schemes will foster tensions that undermine the stability of the bargain so crucial to Prosser’s argument. An example of this risk can be seen in South Africa’s regulatory framework during the period of research. This was an uneasy mix of political and transactional regulation, with the transactional regulation primarily embodied in ‘hard law’ at the national level, and the political regulation appearing as ‘soft law’ or ‘mere policy’. So, for example, in relation to the important power to disconnect consumers for failure to pay, the transactional aspect of securing a stable income-stream for the water providers was the subject of contestation in national hard law that ultimately favoured private providers.³²At the same the government introduced a Free Basic Water Policy in 2001, and a ‘credit control code’ animated by principles of due process and compassion in its 2003 Strategic Framework – important measures, but carefully confined to the arena of ‘soft law’. These soft law and policy dimensions did not, for the most part, reinscribe opportunities for political participation and influence into the regulatory framework in any secure fashion, but focused rather on short-term amelioration of its harshest side-effects. And as some of the judicial consideration of the issues discussed in the case study narrative above show, the protection that did exist was contingent on a conception of a ‘responsible consumer’ that precluded at least the more ‘unruly’ dimensions of political activism.³³

If it is social activism at the local level, including important activism around consumption practices, that performs a reinscription of political agency , how do we integrate ‘unruly’ civil society into a discussion about building bridges between ‘regulatory space’ and ‘citizen space’? On the one hand, routinization is clearly an important dimension of such building. It stabilizes expectations and provides limited predictability, ideally enough to establish a basis for ongoing engagement between actors with diametrically opposed views of how to proceed. This kind of ‘ordering’ does matter – in some of the bitter conflicts that have happened in recent years over the privatization of water services, severe stand-still or counter-productive policy seesaws have arguably been the main outcome, at least in terms of the narrow but vital goal of getting clean affordable water through the taps to the most vulnerable sections of the population.

But an emphasis on routinization should not obscure the substantive political dimensions of water governance. Routinization is not just a technocratic exercise in problem-solving at the margins, but a political process that selectively opens space for some to participate in setting the basic rules and others not to. As a consequence, routinization has important limits: outright disruption and unpatterned conflict are also important in building these bridges, counter-intuitive as that may sometimes seem. Routinization defines itself against the stakes articulated by disruptive protest, and as this paper has indicated, water service delivery in cross-national perspective is often (albeit not always) marked by sustained social protest. The role of disruptive forms of civil society involvement in shaping both the substantive bargains and future procedural routines in the regulatory governance of water service delivery is an important one. What seems illegitimately unruly from a particular perspective or at a certain time can become, from another perspective or at a later time, the prototype of an every-day form of civil society involvement. For example, the civil disobedience campaigns of some of the grass-roots environmental groups in New Zealand evolved into a dialogue with the municipal government that over time became an exercise in mutual problem-solving on the budgetary framework for publicly delivered water services. More generally, the core insight is that the relationships that sustain and legitimate bridges between regulatory and citizen space must be built in each local setting, and that disruptive civil society participation is part, however indirectly, of that process of building.