Archipelagos and Networks: Urbanization and Water Privatization in the South Author(s): Karen Bakker Source: The Geographical Journal, Vol. 169, No. 4 (Dec., 2003), pp. 328-341 Published by: Blackwell Publishing on behalf of The Royal Geographical Society (with the Institute of British Geographers) Stable URL: http://www.jstor.org/stable/3451572 . Accessed: 17/08/2011 16:56

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing and The Royal Geographical Society (with the Institute of British Geographers) are collaborating with JSTOR to digitize, preserve and extend access to The Geographical Journal.

http://www.jstor.org The GeographicalJournal, Vol. 169, No. 4, December2003, pp. 328-341 Archipelagosand networks: urbanization and water privatization in the South

KARENBAKKER Departmentof Geography,University of BritishColumbia, 1984 WestMall, Vancouver,BC, Canada V6T 1Z2 E-mail:[email protected] Thispaper was accepted for publication in February2003

This paper examines the interrelationship between urbanization and water supply privatization in cities in the global South. The purpose of the paper is not to evaluate the impacts of privatization; rather, the paper analyses the differences in pathways and modes of water supply privatization, focusing on urban and contrasting with rural areas. A distinction is drawn between privatization (organizational change) and commercialization (institutional change) of water supply. Emphasis is placed upon the interrelationship between regulatory change (a shift from public to private management of water supply systems), human use of and access to water, and urban waterscapes. In contrast to metaphors of 'networks' so often applied in analyses of water management, the 'archipelago' is posited as a metaphor which better captures the complex overlapping strategies of water supply provision in urban areas in the South. Building on this metaphor, and in response to the 'public-private' dualism often invoked in studies of privatization, the paper outlines an alternative typology of water management in urban areas in the South. This typology foregrounds the concepts of the territorialization of corporate power as a means of understanding the articulation between privatization and urbanization processes in the South.

KEYWORDS: South, urbanization, privatization, political economy, water supply

In Introduction: and 1998). most cities of the global South, private privatization,commercialization, water vendors - water to households by water supply delivering jerrycan or tanker - have long been the means by rivate sector management of water supply which the poor obtain water, usually at a cost per systems is not a new phenomenon. The diver- unit volume several multiples of that delivered via sity of water supply management systems public water supply systems to the middle and worldwide - which operate along a continuum upper classes. between fully public and fully private - bear Despite this history of experience with private witness to repeated shifts back and forth between sector management of water supply, the twentieth private and public ownership and management century witnessed an increase in state dominance of water systems'. In Buenos Aires, a private water (both local and supra-local) of this sector. A combi- company was the first to supply citizens with nation of socio-economic, political and cultural networked water supply. In many cities in France, factors underlies the growing dominance of the private companies manage municipally owned state in water supply provision in industrialized water supply infrastructure via long-term manage- countries in the twentieth century. Water is expen- ment contracts (Goubert 1986). The first companies sive to transport relative to value per unit volume, to supply London with water were privately requiring large-scale capital investments in infra- owned; after a period of state (municipal and then structure networks which act as an effective barrier national) ownership in the twentieth century, the to market entry. Water supply is thus highly suscep- English water supply utilities were privatized by tible to monopolistic control (a condition termed asset sale in 1989 (Bakker forthcoming; Hassan 'natural monopoly' by economists). Commercializing

0016-7398/03/0004-0328/$00.20/0 @ 2003 The Royal Geographical Society Urbanizationand water privatizationin the South 329 water utilities is as a consequence invariably fraught Table1 Waterand sewerageprojects with private with difficulty, to a much greater extent than for participationin developingcountries (1987-2000) other network utility services such as telecommunica- tions, gas, and electricity. Numberof projects Cumulativenumber In addition to water's 'natural monopoly' charac- Year reachingfinancial closure of projects teristics, the symbolic and cultural importance of water as a (partially) non-substitutable resource 1987 2 2 essential for life, its strategic political and territorial 1988 0 2 importance, the intense conflicts that arise over 1989 3 5 the shared use of a flow resource required to fulfil 1990 0 5 multiple functions (agricultural, industrial, drinking 1991 2 7 water, environmental), and the need in industrial- 1992 6 13 ized, urbanized societies to mobilize large volumes 1993 9 22 - invariably at a high cost relative to the economic 1994 15 37 value generated - have been used, particularly 1995 20 57 in the twentieth century, to justify public sector 1996 18 75 involvement. Moreover, the health and hygiene 1997 30 105 effects of lack of access to water, together with 1998 19 124 the tendency of private companies to fail to extend 1999 34 158 coverage to the poor (both as a result of the 2000 25 183 tendency to cherry-pick profitable neighbourhoods and classes of consumers, and the high prices and Source:World Bank PPI database (2001). poor services resulting in a situation of natural monopoly), were two of the most important justifi- cations for bringing water supply under the control tion rather than the rule. Two decades later, the of the state, whether through strict regulation water supply systems of over one hundred cities in or public ownership of water supply infrastructure, developing countries are now managed by one of a during the twentieth century. Thus, throughout handful of private multinational companies eagerly most of the past century, water management, pursuing growth in a multibillion-dollar global particularly but not exclusively in OECD countries, market (Table 1). The projects summarized in Table was characterized by the dominant role of the state 1 include both water resources and water supply. as owner, manager, and regulator of water supply Private companies may be involved in either the infrastructure. construction of large-scale infrastructure develop- Despite the apparent consolidation of state ments, such as reservoirs or canals, to supply raw control over water supply management, the effect- water, or in management, rehabilitation and iveness and legitimacy of publicly owned water extension of 'clean' (i.e. potable water) and dirty supply systems has been challenged over the (i.e. wastewater) networks. The majority of projects past two decades. At the same time, the state undertaken to date involve both 'operation and has increasingly relinquished management and/or maintenance' of the infrastructure,together with the ownership to private capital and ceded regulatory construction and/or rehabilitation of infrastructure. control to markets and market-mimicking regula- In the case of water supply, private sector participa- tory mechanisms through a combination of tion may take a variety of forms (Table 2). deregulation and re-regulation. Argentina, Bolivia, To date, the majority of contracts have been China, Chile, England, Indonesia, Morocco, the granted on a 'concession' basis, in which a private Philippines, Poland, South Africa, Spain, Thailand, company obtains the exclusive right to operate the and Turkey are just a few of the countries in which water supply infrastructure for an extended period the state has initiated private sector participation in of time (usually 20-30 years). The vast majority of water supply2. The rapid increase in private sector contracts have been granted in urban areas; given participation has taken a variety of forms: the economies of scale associated with concession opening of the water supply industry to the private contracts, rural areas, or even conurbations with a sector in the majority of EU countries (Petrella population less than 500 000, are unlikely to attract 2001); private-public partnerships in the South the interest of the formal private sector. (Silva et al. 1998); the creation of water banks and The majority of these large projects have been markets (Bauer 1998; Giansante etal. 2000); and undertaken by a very small number of companies. the massive international growth of the bottled Local private companies tend to partner with the water market. At the beginning of the 1980s, the sector leaders - ONDEO/Suez Lyonnaise des Eaux private management of water supply was an excep- and Vivendi/G6nerale des Eaux (both French) - both 330 Urbanizationand water privatizationin the South

Table2 Privatesector participation in watersupply - contracts

Contracttype

Service Management Allocatedresponsibility contract contract Lease BOO, BOT* Concession Divestiture

Assetownership Public Public Public Publicand Public Privateor private publicand private Capitalinvestment Public Public Public Private Private Private Commercialrisk Public Public Shared Private Private Private Operationsand Publicand Private Private Private Private Private maintenance private Tariffcollection Public Public/private Private Public Private Private Duration 1-2 years 3-5 years 8-15 years 20-30 years 25-30 years Indefinite(may be limitedby license) Examples Mexico City Gaza City Guinea Sydney BuenosAires London

*BOOstands for 'buildown and operate'and BOTfor 'buildoperate (or own) and transfer'

Fortune 100 companies; on most measures, these brought to the table . . . [World Bank investment] is two companies control approximately 70% of the critical in the privatesector feeling secure. market. This concentration is due in part to the World Bank/IFC1991, 4 absence of private sector local competitors, and in part to the competitive advantage held by these The facilitation role played by both international multinational firms, collectively serving hundreds financial institutions (IFIs)and national governments of millions of customers on all five continents. is a critical and necessary condition for the existence Another advantage has been the fact that private and functioning of the world water market. In the companies receive support from key mediators of Far East,where 'water privatisationhas been promoted international finance - the International Finance by IMF pressure' (British Water 1998, 3), agencies Corporation, the World Bank, the African, Asian such as the IMF encourage receptiveness for and American Development Banks - and key private involvement in the water sector in 'client' sources of bilateral aid, such as Britain's Depart- countries, facilitate contacts between 'developed' ment for International Development. world water companies and overseas govern- Given the limited number of competitors, ments, and provide financial assistance and markets must be stimulated and simulated through guarantees for water privatization projects. The role making 'conditions as favourable as possible for of these international facilitators is crucial. For the [companies] to compete [and through] finding example, bilateral aid agencies such as Britain's ways of increasing the number of interested com- Department for International Development provide panies, both internationally and locally' (World technical assistance for countries considering Bank/IFC 1991, 5). Markets must also be created privatization. In some developing countries, IMF and fostered through the promotion of private conditionality requiring privatization and/or com- sector involvement (e.g. through 'seed' loans and mercialization of the water supply sector has been grants for technical assistance in preparing and imposed3. implementing concessions and leases), and the use The current era of water privatization can be of guarantees to promote private sector involve- distinguished from previous eras by four character- ment (World Bank 1997b, 46). More subtly, the istics. First, the scale of involvement of multi- World Bank realizes it plays a key role in creating utility, multinational companies is unprecedented. investor confidence: The two largest companies, ONDEO/Suez Lyonnaise des Eaux and Vivendi/G6nerale des Eaux (both an institutionwhich 'brings confidence', such as the French), are among the largest 100 companies in Bank, might have influence on the psychological the world, and dwarf their few credible competi- factorswhich may be much more importantthan money tors. Second, the amounts of finance being Urbanizationand water privatizationin the South 331 mobilized, via these private companies as well as subsidy to poorer consumers. Privatization can multilateral lending and bilateral aid agencies, are occur without liberalization; the divestiture of large and growing, particularly in response to rapid the public water companies in England and urbanization and a consequent reduction of access Wales in 1989, for example, preserved their to sufficient supplies of potable water for millions vertically integrated, monopoly structure. Com- of city dwellers worldwide. The availability of petition for the market may occur without private finance must be understood within the context of ownership of infrastructure; in France, municipali- increasing flows of foreign direct investment to ties own water supply infrastructure, and may the South, beginning in the late 1980s. Infrastructure choose to tender water services out to private investments by the International Finance Corpora- companies who compete for long-term contracts. tion (a World Bank affiliate that lends only to Privatization and commercialization may thus be private entities), for example, surged 'from modest distinct modes of transformation of water services amounts in the late 1980s to $330 million in the management. fiscal year 1993 leveraged more than ten times In practice, however, private sector participation so that, in 1993, the IFC participated in private and/or privatization of water supply often imply [infrastructure] investments of $3.5 billion' (World commercialization. For the proponents of privatiza- Bank 1994, 92). Third, as discussed above, private tion this is of clear benefit; water must be treated companies are receiving both ideological and as an economic good - as specified, for example, financial support from key mediators of interna- in the Dublin Principles4 and in the Hague tional finance. Fourth, water privatization is Declaration5 - if it is to be managed efficiently and occurring as part of a debate over the sphere of if greater numbers of people are to be provided legitimacy of the state, and a much broader process with access to safe, sufficient water supplies. of delegation of formerly core state functions to Countering the claim that water is an economic non-state actors, in the context of high levels of good (and a human need), the Water Supply and state indebtedness. Sanitation Collaborative Council's 'Vision 21'6, the It is within this context that utility services such Cochabamba Declaration7, the Group of Lisbon's as telecommunications, electricity, and gas, and Water Manifesto8, and the UN Committee on infrastructure such as roads and housing have Economic, Social and Cultural Rights' statement9 been increasingly privatized and commercialized on the right to water have argued that water is a over the past two decades. Privatization refers to human right under international law. the shift in ownership from the public to the The question of whether water supply is a private sector. Private sector participation entails human right or a human need bears directly on the the participation of private companies and private water privatization debate. In the mid-twentieth capital, through a variety of contractual arrange- century, international debates stressed the import- ments to build and manage infrastructure on behalf ance of water for health and sanitation in basic of the public sector. Commercialization refers to a need requirements. In recent decades, the argu- reworking of the management institutions (rules, ment for treating water as a human right has norms, and customs), and entails the introduction been advanced, drawing on the Universal Declara- of markets as allocation mechanisms, market- tion of Human Rights (1948), the Covenant on simulating decisionmaking techniques, and the Economic, Social and Cultural Rights (1976), and displacement of Keynesian-welfarist by neo-liberal as made explicit in the Convention on the Rights principles in policymaking. A key element of of the Child (1986). Water as a human right would commercialization of services is liberalization: be enshrined in legislation (as in South Africa's selective de-regulation and re-regulation designed constitution1o) placing a duty on governments to to allow and indeed encourage competition in ensure the fulfilment of this right. If water were a the product market. Direct competition in the water human need, however, governments would have market is in most cases unfeasible, yet various no such duty. In practical terms, a human right to forms of competition, for rather than in the market, water would imply a basic volumetric allocation and of simulated competition have been intro- per person per day; 'sufficient for everyone's need, duced (Cowan 1997; Sawkins 1995). but not for everyone's greed'. The difficulties of Privatization and commercialization are not implementing such a right are well understood in necessarily concomitant. Some publicly owned South Africa, where many citizens have been water companies (e.g. Amsterdam's) are run along promised, but not yet provided with, the minimum fully commercialized lines. The converse can also level of 50-60 litres per person per day established be found, most frequently in the case of public- by the government as 'sufficient' in its post-apart- private partnerships in developing countries which heid Reconstruction and Development Plan (Bond make concessions in the form of direct or cross 1998). 332 Urbanizationand water privatizationin the South

Water and sanitation in urbanizing areas Expanding the network poses distinct technical difficulties, depending on the spatial distribution, We are in an era of giganticism,where nation-wide physical layout, and concentration of housing. In economic survival depends on gigantic organisations most cities, a networked water supply system exists for mass production.We no longer irrigateour own but covers only a proportion of the city, usually small-holdingsand water our cattle from the village part or all of the most affluent neighbourhoods. pump; we mass together for work, we mass-produce Expansion usually implies extension of services to in mammothfactories.. . and therefore,we must mass newer, often informal settlements in urban areas. produceour water. This requires an enlargement of existing and build- Twort1963, 9 ing of new trunk mains, water treatment and wastewater treatment plants. In many cases, new water sources such as reservoirs - and Urbanization and the industrialization of water expensive time-consuming to develop - are also required. supply production A second set of difficulties relate to institutional Water lubricates capital circulation, a necessary, factors. In a weak institutional setting, for example and yet often invisible precondition to life in in the absence of secure land tenure, it may be modern, industrialized societies (Swyngedouw difficult to obtain the necessary permits to build 1997). The emergence of an industrydedicated to the infrastructure. Poor quality information - cadastral mass production of water, and the concurrent surveys in newer, and particularly informal settle- conversion of water from an artisanal to an indus- ments are often lacking - confounds the creation of trial product, tends to occur in urban areas during a potential customer base and renders cost recov- periods of rapid urbanization. In rapidly urbanizing ery more difficult. Communities often have strong areas, the limited availability of clean water internal organizational structures, but these groups supplies leads to widespread concerns over water may not have formal status and may not be recog- quality and access, as artisanal water sources - nized by the water company or the municipality as wells, rivers, ponds, streams - become increasingly legitimate interlocutors. polluted (Lindh and Gilbrich 1996). To meet In addition to technical difficulties and institu- growing demand, water production becomes tional weaknesses, a logic of rent-seeking and gradually industrialized, with artisanal methods of patronage on the part of urban elites often under- collecting and distributing water (such as water lies the failure to expand urban services. vendors moving on foot through city streets) being Governments may be unwilling to extend services replaced by networks of pipes leading from reser- to informal settlements due to their unwillingness voirs to consumers' taps, achieving economies of to support or encourage additional rural-urban scale and scope, and enabling higher consumption migration. The existing public water supply levels. Surface watercourses - which typically network is frequently badly managed; sufficient support a multiplicity of uses including transport, water resources may exist, but leakage and high trade, drinking water, and effluent disposal - are rates of unaccounted for water may significantly filled in, or covered over. The water network is, in reduce the availability of water supply. In many part, an artefact of urbanization. cases, the public water supply network operates in a vicious cycle of the '3 lows': low investment, low service standards, low cost recovery. In some The urban services challenge cases, corruption and the misuse of the water There are numerous difficulties posed by the provi- supply business for political ends occurs; water sion of services such as water and sewerage in services are particularly open to such behaviour rapidly expanding and unplanned urban zones. as public water companies are one of the few Technical difficulties may be posed both by a rapid revenue-generating utility services controlled at increase in demand (volume of water and waste- the municipal level. As such, they are frequently water), as well as the distribution of demand. employed to subsidize other municipal services Water and sewerage infrastructure networks are and activities, without applying the surplus (if any) highly capital-extensive, with relatively long lives. to network maintenance and expansion. Rent- They are also sunk networks - often physically seeking and patronage aside, the multiple demands buried below city streets and, in many cases, on water supply utilities in urban areas - where below layers of other infrastructure networks. Built industrial and institutional sectors account for to peak load specifications that may have become the majority of water use, and are well placed to quickly outdated, water and sewerage networks are advance their claims for services - also contribute highly inflexible - very expensive and unwieldy to to the relative neglect of some residential update. neighbourhoods. Urbanizationand water privatizationin the South 333

This behaviour must be understood in the context out the developing world (see, for example, Lovei of 'urban primacy', the pattern of dominance of and Whittington 1993). one (or, at most, a handful of) urban area(s) within It is in the context of inequitable access to water many, if not most, 'developing' countries. As Gilbert supply that calls for improving water and sanitation and Gugler note, in many third world countries, services must be viewed, particularly in the context 'most large-scale modern activities, most forms of of rapid urbanization in the 'global South' (see, social infrastructure, and most centres of decision for example, Kjelln and McGranahan 1997). Annual making are found in a single major city' (1992, 36). investment in urban water supply in Africa, Asia Unlike the feudal order of Europe's Middle Ages and Latin America and the Caribbean over the or the 'hacienda' system in the colonial and early decade 1990-2000 was just under US$8 billion post-colonial period in Latin America, throughout (half funded by external support); annual invest- the South today virtually the entire elite is located ment in rural water supply was significantly in cities (Gugler 1997). The social surplus that smaller, at just over US$4.5 billion (with external would have been controlled by private interests support funding just under half) (WHO 2000). In under laissez-faire capitalism, or by colonial powers the context of government investment, these figures during the age of empire, is now controlled by are significant: the medium total investment in the state, which reallocates resources with three water supply and sanitation as a percentage of goals: improving the immediate environment (of overall government investment during the same decisionmakers); 'assuring the continued collabora- period was 5.3% in Africa, 3.6% in Asia, and 8.3% tion of the middle class'; and 'placating in Latin America and the Caribbean. strategically placed elements of labour... with the The imperative to change this pattern of unequal result that public resources are disproportionately distribution of potable water has lain at the heart spent on the privileged consumption of the few, of calls for increased expenditure on water supply and conspicuous investment for the few - in cities' and sewerage services since the International (Gugler 1997, 120). Within this hierarchical, differ- Water and Supply Sanitation Decade in the 1980s entiated socio-political model, individuals do not (WHO 1990). Water supply and sanitation exp- have equal entitlements to state services by virtue enditures, as a proportion of total aid (or of citizenship. Not all citizens, in other words, are 'cooperation') budgets in OECD countries, have considered to be political constituents of society. increased steadily since the 1980s to 6.6% in 1996 Rather, urban elites - with entitlements to local (WHO 2000). However, in the decade from 1995 state services such as piped water supply - consti- to 2005 the World Bank estimates that US$60 tute the privileged constituency of the local state. billion per year must be invested in the sector to Indeed, urban elites are often state elites, staffing ward off a situation of 'water stress'" (Haarmeyer state institutions and constituting a core base of and Mody 1997): political support for governments. Currenttrends show that several regions, most notably the Middle Eastand North and an Water and sanitation services in areas of Africa, increasingly urbanizing numberof countries in all the South large partsof the world are approachinga 'watercrisis'. For the urban elite, water supply is often relatively World Bank 1997b, 47 abundant, and relatively cheap. For the urban poor, the scarcity of potable water is a daily The increase in 'water stress' or 'water scarcity' as hardship. Wealthy, mostly white South Africans, for measured by indicators linked to unit volume avail- example, use on average 600 litres per person per able per consumption masks a more profound, day of potable water delivered through taps inside widespread, and longstanding experience of water the home, whereas the poor and largely black scarcity in cities. On a global scale, consumption residents of peri-urban settlements may average 10 of water by domestic users is only 4% of total litres per person per day (Van der Merwe 1995), water consumption. Yet twenty-first century urban often walking several kilometres to fetch water. dwellers consume large quantities of water; each This highly unequal distribution of water is in this inhabitant of London uses perhaps 160 litres per case the legacy of apartheid, which entrenched person per day, and the inhabitants of Los Angeles differential access to water resources and domestic closer to four times that amount. In industrialized water supply networks (Bakker and Hemson 2000), countries, demand per capita has increased steadily but is by no means unique. The pattern of elite throughout the twentieth century. Whereas wealthy access to piped water supply, and reliance of the residents of desert states such as Nevada consume poor on the informal water sector (water vendors) on average over 950 litres per person per day or unimproved water sources is repeated through- (Solley et al. 1998), the World Health Organization 334 Urbanizationand water privatizationin the South estimates that 1.1 billion people worldwide do not tional response to the problem of providing access have access to safe drinking water, and 2.4 billion to water supply was an argument in favour of are without access to adequate sanitation (WHO increased state spending. In addition to the multiple 2000). market failures'2 characterizing its supply, the Investment in urban areas is directed in part uniqueness of water - as a partially non-substitutable to meet the water supply and sanitation needs resource essential for life and critical to public of rapidly increasing urban populations. Cities health - was a strong justification for managing globally have experienced since 1950 what Harvey its supply as a service, which should be available terms 'hyber-urbanization', with 'the pace of on a subsidized basis to citizens. Water supply urbanisation accelerating to create a major ecolog- was perhaps the most basic of entitlements, and ical, political, economic and social revolution in running water inside the home considered to be the spatial organisation of the world's population' a powerful material emblem of citizenship. More (1995, 10). Between 1950 and 1985 the proportion pragmatically, given the high degree of public of the world's population living in urban areas health externalities, water supply was reconceived doubled. By 1985, in the more affluent countries in as a 'merit good'3 - necessary for both production the Third World urbanites had become a majority. and reproduction. After a period of experimenta- As cities have grown, urban services have not tion in the nineteenth century in many countries kept pace. The number of people without access with private sector provision of water supply, the to sanitation in urban areas was estimated to have local state increasingly dominated water supply, increased by 10 million in the 1980s (Haarmayer and any remaining private activity was, in most and Mody 1998). If the rate of urbanization does cases, strictly regulated. not decrease significantly, this number is predicted In contrast, arguments in favour of increased to grow (Haarmayer and Mody 1998). The World private sector participation attribute the failings Bank estimates (lower than those of the WHO) that of water supply management not to the failure to 1 billion people have no access to safe drinking guarantee entitlements, but to the assumption that water, and 1.7 billion are without access to adequate entitlements are a mechanism by which potable sanitation (World Bank 1997b). An increasing water can be supplied: proportion of those lacking potable water live in urban areas. In most developing countries, access in the majorityof countries. . . water has been treated is highly correlated with income. as though it were available in unlimited quantities, Even those with access to networked services and supplied at zero or low cost to consumers who may have problems with low quality and reliabil- resent the idea of water as an economic resource. ity. In many networked water supply systems, Consumers,abetted by their governments,have clear water may flow intermittently - a few hours per notions of their water 'requirements,' and the task day, or only a few hours per week. Water pressure of water authorities has, until recently, been seen may be low, further reducing access. In some as supplying those needs, with cost a secondary households, a water tap is left constantly open - consideration. Pricing for water services has been when the arrival of water announces itself in a meagre and sporadic, and is normally incidental to sudden gush from the tap, users rush to fill buckets cost-recovery, narrowlyconceived. This 'entitlement' and other vessels with water. In part because of syndrome, relying on supply-side solutions to problems associated with infiltration in pipes in requirementstaken as given, is unsustainable. which supply is intermittent rather than constant, Winpenny1994, 2 water is frequently not potable, and may be of insufficient quality for other household uses, such The reference to 'unsustainability' refers, in as cooking. As a result, most households rely on a this case, to economic sustainability: the ability mix of water supply strategies: for the wealthy, a to recover costs in order to meet operations and tank on the roof connected to both a private deep maintenance requirements, and eventually fund well and the network, supplemented by bottled network extension costs. The inability of many water for drinking; for the less affluent, a hand- local states to mobilize revenues from users of the dug shallow well for bathing and cleaning, often existing water supply system - frequently expressed in conjunction with a supply of drinking water as very low cost-recovery rates - is one of the purchased from neighbourhood water vendors. widespread failings of municipally owned water utilities in developing countries. According to the 'entitlement' has Privatization and commercialization of water supply Winpenny, syndrome additional 'perverse' economic effects, such as management in the South the encouragement of water-intensive industries, Over much of the twentieth century, the conven- the encouragement of inefficient uses of water, Urbanizationand water privatizationin the South 335

discouragement of technical innovation necessary commercialization are thus often accompanied by, for conservation, and under-investment in water and implicated in, a more generalized shift in the infrastructure due to low revenues and consequent practice of deliberative democracy at the local low profitability. level. The initiation of private sector participation In essence, neo-classically derived arguments in contracts, for example, is often accompanied by favour of valuing water rest on the assumption that a discursive rescripting of users as individual 'the failure to treat water as a scarce commodity consumers, rather than a collective of citizens, and lies at the heart of the [water industry's] problems'14 by a higher prioritization of environmental water (Winpenny 1994, 1), most frequently identified quality standards. Yet opponents of privatization as water pollution, over-abstraction, and under- and commercialization often object to this alliance investment. This failure, caused by the (inter- between commercialization and conservation. As a related) factors of externalities, a 'political' support response to the Hague Declaration, the Declaration for the notion of water as a public good, and the of the P7'5 at their fourth Summit in 2000 outlined under-pricing of water, is exemplified by subsidiza- principles of 'water democracy': decentralized, tion of water production and non-volumetric community-based, democratic water management consumption charges. in which water conservation is politically, socio- The above argument rests upon the assumption economically and culturally inspired rather than that flawed management by the state, due to struc- economically motivated (see also Shiva 2002). tural defects in public sector management of water, Rather than defending the state, water activists is responsible for the well-documented poor quality are increasingly outlining alternative visions of and low penetration of water supply systems. community-centred resource management (see, The state, argue organizations like the World for example, IFG 2002). Bank, is 'overextended' (World Bank 1997a); These debates have increased in scale and scope only by 'relaxing the government's grip' can over the 1990s, in part in response to the dramatic countries 'free up public resources for high- increase in private sector participation in the priority activities; pave the way to better, cheaper financing, construction, and management of water services; and unlock opportunities for private supply infrastructure. During the last 14 years of sector development' (World Bank 1997a, 61-2). the century, 183 water and sewerage projects with The underlying assumption is that the market is private participation were initiated in 'developing' more efficient than government at providing basic countries (Table 1), with a total investment of over services. Given 'state failure', 'there is no good US$33 billion (World Bank 2001). The water sector economic reason for state ownership to persist in is not unique; capital for infrastructure investment tradable-goods industries' (World Bank 1997a, 64). was increasingly sourced through the private sector Underpinning this statement is a two-pronged during the 1990s. Capital may be raised by the discursive move. First, the category of tradable state through leasing or divesting infrastructure to goods has been expanded, and the utility sector private firms, and these same private firms - who simultaneously re-conceptualized as potentially are assumed to have better access to loan finance profitable, rather than a service provider in need of than governments - are also contracted to operate, subsidies. Second, water scarcity is depicted as a and in some cases build additional infrastructure. universal condition; simultaneously natural, justifying The move to the private sector began in the 1980s; cost-reflective pricing and the commercialization between 1988 and 1992 alone, US$1.6 billion in of water, and social, the result of flawed public revenue was obtained by 'developing' countries management. Both depictions of scarcity may serve from the privatization of public enterprises, one- as a further justification for water privatization and third of which came from the privatization of infra- commercialization. structure entities (World Bank 1994, 105). Proponents of water privatization in both cases These capital flows are being mobilized to stave rely, if not always explicitly, on the concept of off what the World Bank portrays as the possibility state failure - the notion that the private sector is of 'severe water shortages' in developing countries, more efficient than the public sector at delivering and to meet tougher environmental standards in basic services - supported by evidence from developed countries (McGuinness and Thomas rapidly urbanizing Third World cities, where most 1997, 329). A significant proportion of this finance, of the estimated 1 billion people who lack daily insist multilateral credit organizations like the access to sufficient amounts of clean water live World Bank and bilateral agencies like DFID, must (WHO 2000). Their arguments converge with come from the private sector; development banks those who promote the management of water as and other funding institutions are 'increasingly an economic (rather than public) good in order prescribing the introduction of Private Sector to promote water conservation. Privatization and Partnerships as a condition of lending' (British Water 336 Urbanizationand water privatizationin the South

1998, 11). The level of investment in water supply is pay a reasonable price for an appropriatelevel of likely to increase, given a combination of increasing water and sanitation services; the willingness to pressure on water resources, and increasing pay has to be developed prior to and during [the demands, particularly in conurbations experiencing introductionof] Public PrivatePartnerships. rapid rates of urbanization. DFID1998, 10

The underlying assumption in the DFID argument Commercialization of water supply management is that full cost recovery pricing is both possible Although privatization and commercialization are and desirable. Users should no longer be charged not dependent variables, private sector participation on the basis of the 'ability-to-pay principle' - the frequently entails some degree of commercializa- principle which typically underpins water pricing tion. Whether through a reworking of infrastructure in public, non-commercialized systems which management goals, or through a redefinition of implies that users should be charged according to principles underlying the business of water supply, their ability to pay. Rather, users should be charged water ceases to be a service, supplied at subsidized on the basis of 'economic equity' - the principle rates to citizens as a right, and is increasingly that users of a utility service should pay, as near as viewed as a commodity, sold to consumers on possible, the costs they individually impose on the a profit-making basis of willingness-to-pay rather system (the 'benefit principle'). In other words, a than ability-to-pay. commercialized system which adopts a 'full cost Commercialization is often justified by the recovery' pricing policy is the preferable solution 'perversity' of public sector management, which is to the water supply needs of the urban poor. The most forcefully revealed, as it is argued by private unspoken caveat is that inter-generational equity is sector participation advocates, in the effects of not considered; no redress is possible for the past water subsidies, which 'almost always benefit the subsidization of the typically 'non-poor' at the non-poor disproportionately' (World Bank 1994, expense of the poor. Whereas the connected 80). The under-pricing of water by governments segment of the population - typically more affluent in order to keep water tariffs low is a widespread - enjoyed both capital and operating expenditure policy in both 'developed' and 'developing' subsidies, the poor must pay the full cost of countries. In the latter, however, low penetration connection and supply. rates imply that only a small, and typically a In practice, this recommendation has not proved wealthier-than-average proportion of the popula- to be workable, given the low ability-to-pay of tion are connected to the public water supply many domestic water users in the South. Post- system. The poor typically rely on private water privatization, a continuation of cross-subsidies vendors or public standpipes and 'typically end frequently occurs. In Buenos Aires, for example, up paying much higher prices for infrastructure where the water supply and sewerage systems services or their substitutes' (World Bank 1994, were privatized by concession in 1993, an initial 81), in some cases indirectly subsidizing the public pricing scheme which charged the full cost of water supply system (Swyngedouw 1997). The connection to new users was discontinued after solution is to enlarge the public water supply widespread non-payment or refusal to connect was system - but not necessarily to decrease the prices found to be occurring in poorer zones of the city. paid by the poor. Connection charges were replaced by a blanket By contrast, the high prices paid by those relying charge to all household connections, with the on the 'informal' water sector are interpreted as proceeds intended to fund network expansion. In evidence of their 'ability' (or sometimes, and more Chile, water users receive a bill for the full cost of dangerously, their 'willingness') to pay for water the service, but poor households can obtain a services, at rates high enough to ensure cost recov- partial rebate on water charges via the municipality ery, and even profitability of water supply systems. (Serra 2000). In South Africa, an initial attempt at As one DFID publication notes: implementing full cost recovery pricing policies - despite the constitutional guarantee to 'sufficient People in poorer areas where piped supplies are water' - stalled after large cholera epidemics. A not available usually have to buy their water from 'lifeline' water supply policy, in which a pre-set vendors. Experiencearound the world shows that the volume of water is provided at minimal or no cost cost of buying water in this way is far higher, from a to poor users, has now been implemented, but the low of 4 times up to 100 times more than from public question of allocation of capital costs, and operat- utilities. This mostly affects the poorer members of ing and maintenance expenses remains unresolved society who are least able to affordto pay - but pay in many communities (Bakker and Hemson 2000; they must, in order to survive. The ability is there to Bond 1998). Urbanizationand water privatizationin the South 337

Urban and the territorializationof mechanisms of control of water production waterscapes co-exist 1). In most cities in corporatepower frequently (Figure the South water supply networks do not operate Water as a resource circulates through the hydro- homogeneously over the urban landscape; they social cycle - a complex network of pipes, water will overlap with, and be inter-penetrated by, alter- law, meters, quality standards, garden hoses, native service delivery mechanisms. The metaphor consumers, leaking taps, as well as rainfall, evapo- of the 'archipelago' - spatially separated but linked ration, and runoff. The water supply 'network' thus 'islands' of networked supply in the urban fabric - extends far beyond the pipes conveying water to perhaps better captures this complex layering of customers' taps; it is not bounded by the physical use-values and modes of production. infrastructure that abstracts, treats, and distributes Within urban areas, both corporate"1 and non- water, and removes wastewater. Exchange relation- corporate organizations are involved in the provi- ships, demand patterns, customers' expectations sion of water supply; this is unsurprising, given that about water quality and pressure, laws at national corporate modes of organization tend to be created and supranational levels concerning water quality, when water supply is industrialized and extensive rainfall patterns, even climate change shape the water supply networks are created. Simultaneously, flow of water through the pipes. Water circulation, small-scale, unincorporated private (often informal) in short, is dependent upon institutions and water supply businesses operate - particularly in practices as much as on the hydrological cycle; it informal settlements and poor neighbourhoods. is not only socially produced, but also socially Community water supply cooperatives, in which enacted (La Porte 1994, 271). users are typically involved in governance as well The fixed capital embodied in the material as operations and maintenance of water supply network of mains, pipes, pumps, reservoirs, and systems, are also relatively common in urban areas. sewers can thus be viewed as an artefact of the The public-private dualism thus breaks down when hydro-social cycle, shaped by successive genera- attempting to analyse water supply provision in tions of social, technical, and economic practices. cities of the South. 'Privatization' is better read as As an artefact of urbanization, water supply an overlapping set of strategies - industrialization, networks in cities in the North embody the succes- corporatization, and internationalization - of sive phases of industrialization and corporatization water supply in zones where a high degree of non- of water management. In cities in the South, in corporate activity already exists. Not only do contrast, networks are often partial; water and management responsibility and/or ownership move wastewater are metabolized through a complex from the public to the private sector; as depicted temporal and spatial weave of water use practices in Figure 1, the mode of production of water supply and methods of disposal. Industrial and artisanal is transformed from artisanal to industrial (i.e. modes of production, state and private sector network infrastructure) production. Institutionally, ownership, collective and corporate institutional the experience of water supply provision for urban

ARTISANAL Private for-profit Wells, Water ,actors rivers, Ivendors streams Bottleded COMMUNITY Watercoop water CORPORATE CONTROL CONTROL

Municipal State water water corporation Water service MNC

I INDUSTRIAL Figure1 Modesof urbanwater supply provision 338 Urbanizationand water privatizationin the South

residents undergoes a transition from community to water supply privatization taking place, the urban corporate control. The multi-dimensional nature of fabric is being homogenized, with universal service these changes has led some water managers in the provisions backed up by the state as regulator South to refer to the transition underway in water (however imperfect). The 'retreat of the state' in this provision as a 'transformation', rather than merely case deserves close re-examination, as a partial 'privatization' (Inglese 2002). retreat and partial expansion of local state activity, This 'transformation' is most acute in socio- and also as a highly ambiguous process for the economically marginal urban areas, in which urban poor, who may have fewer nominal entitle- urbanization both necessitates and drives the ments as citizens, but who are promised greater industrialization of water production and cor- claims for substantive entitlements. poratization of control of water sources. The This process should be clearly distinguished from displacement of artisanal activities (themselves 'privatization' in rural areas, which often entails sometimes highly exploitative) by provision via the dismantling of relations of mutual cooperation corporate-controlled networks implies the formali- amongst water users, enacted through juridical zation of the enclosure of the hydro-commons, and and regulatory mechanisms that permit (and often the territorialization of state power - whether as encourage) private individuals and corporations regulator or owner of the water supply infrastruc- to capture formerly community-controlled and ture. This is often accomplished by the physical managed water sources (Orellana forthcoming). burial of watercourses, regularization of legal land Just as in urban areas, privatization in rural areas tenure and cadastral records, creating a series of entails a shift in power to, or amongst, local elites. (occasionally overlapping) monopolies through the In rural areas, however, given the lack of an infra- containment of accessible water in underground structure network and economics of scale and networks. Regulations prohibiting self-provision environmental conditions that mitigate against frequently accompany this transformation. In building networks, privatization does not usually the cities of Jakarta (Indonesia) and Cochamba result in the industrialization or internationalization (Bolivia), where private sector concession contracts of water supply provision, which are far more were signed in 1997 and 1999 respectively, priva- likely to apply to water resources production for tization occurred together with a change in primary extractive and secondary industries. In municipal bylaws establishing the right of the state Bolivia, for example, privatization of water supply to regulate groundwater abstractions, and the right by international companies has been restricted of the private company to charge for abstractions to larger urban areas (the cities of La Paz and from private wells within the concession bounda- Cochabamba). In rural areas of the country, in ries (Braadbart 2001; Crespo 2002; Laurie and contrast, international water companies have focused Marvin 1999; Marvin and Laurie 1999). In the case on water resources with export potential - for sale to of Cochabamba, some attempts were made to Chilean mining companies, for example7. enforce these provisions of the contract, contribut- to the resentment of ing local users and to the Conclusions strength of the broad-based coalition of peasants, farmers, organizer labour, environmentalists and Whether in rural or urban areas, water supply women's groups, which succeeded in forcing the privatization redraws the hydro-social landscape. government to cancel the private sector participa- Urbanization, in many cases, facilitates the indus- tion contract. trialization of water supply, enabling the creation of As the above anecdotes demonstrate, commer- the networks that in turn enable corporate control, cialization is a mechanism whereby the state whether public or private. Given the economies progressively expands - in a spatial and institu- of scale required to attract private sector investment, tional sense - regulatory authority. In urban areas in the vast majority of private sector participation the South, the local state in the past typically failed contracts in water and sewerage currently operate to extend public services to socio-economically in urban areas. Private sector participation projects marginal areas of the city, for reasons discussed in rural areas pose very different technical and in the section 'Water sanitization and urbanizing institutional challenges (Bakker and Hemson 2000). areas'. The result, as argued above, was that state We should not be speaking of water privatization power did not operate continuously over the in general, but of the privatization and com- urban fabric, but rather in the case of public mercialization of either urban or rural water services services was constructed as an 'archipelago' highly in particular. correlated with socio-economic status. With the This paper has attempted to highlight the retreat of the local state from a role as operator urban dimension of water supply privatization and of public services in the context of a new phase of commercialization in several ways. First, this paper Urbanizationand water privatizationin the South 339 has argued that 'privatization' is in many cases a Acknowledgements misnomer, given the diversity of water use provision comments were received from Bill and in urban areas. 'Privatiza- Helpful disposal practices Dorman, Alex Loftus, Philippe Le Billon and three tion' is better read as an set of overlapping anonymous reviewers. Presentation at the DARG - industrializationof water strategies supply production, session on 'Landscapes of change' at the IBG the territorialization of in zones special corporate power conference in Belfast (2002) the original where a of provided high degree non-corporate activity already impetus for this article; the encouragement of exists, and the internationalization of control of Chasca and Frances Harris was crucial to water Twyman supply. its completion. Water supply privatization is from this perspec- tive understood to be one dimension of a broad- Notes based process of transformation - institutional, social-economic, technical - as both a response to 1 This is more true for water supply (i.e. water supplied in the needs of capital and as a mechanism of reticulation networks) than for water resources (i.e. bulk the redistribution of power amongst local elites. water). There are many more examples of water supply pri- This 'transformation' is better understood as three vatization than resources privatizationand commercializa- inter-related shifts in water supply management: tion, Chile, the southwesternUnited States, and the Canary industrialization of water supply production; corpo- Islandsbeing the most frequentlycited examples. ratization of water supply management; and 2 For references on 'developing' countries, see, for example, internationalization of water supply control and Batley (1996), Blokland et al. (1999), Cook and Kirkpatrick regulation. The experience of this 'transformation', (1998), Franceys(1997 2000), Johnstoneand Wood (2001), particularly for the urban poor, is of the interre- Lee (1995 1996), Nickson (1997), Rivera 1996, Roger lated processes of increasing territorialization of (1999), Rondinelli and Cheema (1988), and Shirley and corporate power (both public and private), enclo- Walsh (2001). sure of the hydro-commons, and the increasing 3 For an analysis of the 'cross conditionality'which has been penetration of the interests of (largely urban) elites imposed on the water sector in several countries, see into not only rural areas but also peri-urban and Grusky(2001). economically marginal urban areas. This change 4 The 1992 International Conference on Water and the is more multi-dimensional than privatization Environmentset out what became known as the 'Dublin processes in the North, which tend to occur in Principles':Fresh water is a finite and vulnerable resource, urban areas in which high levels of network essential to sustain life, development and the environment; penetration exist, and where the intervention of the Water development and managementshould be based on a state in the urban fabric is more spatially homoge- participatoryapproach, involving users, planners and poli- nous. The metaphor of the 'network' is widely cymakers at all levels; Women play a central part in the employed by water supply managers, multilateral provision, management and safeguardingof water; Water credit agencies, and development organizations has an economic value in all its competing uses and should alike, but is rooted in a Northern bias. A metaphor be recognized as an economic good. The Dublin Principles that better captures the experience of urban water have been adopted by numerous international,multilateral supply in much of the South is the 'archipelago'. and bilateralagencies includingthe World Bank. Whether privatization will, as its proponents 5 The MinisterialDeclaration of the Hague on Water Security promise, convert archipelagos into networks remains in the twenty-first century followed the inter-ministerial to be seen; to date, there are few examples of meeting known as the '2nd World Water Forum' in 2000. private sector participation contracts which have See http://www.worldwaterforum.net. systematically countered the spatial differentiation 6 The Water Supply and Sanitation Collaborative Council, of water supply access in urban areas. Privatization located in Geneva, is a non-profitorganization that acts as thus remains a highly ambivalent process for urban an 'internationalpolicy think tank' on water management. water users, particularly the poor, given the associ- 7 The Cochabambadeclaration followed a meeting of several ated reworking of substantive and procedural hundred people in this Bolivian city concerned about entitlements to water supply. Analyses of water the involvement of private sector corporations in water privatization in cities in the South must address supply management. See http://www.canadians.org/blue- this multi-dimensional transformation, in which the planet/cochabamba-e.htmi(Barlow and Clarke2002). involvement of the private sector in water supply 8 The Group of Lisbon is a group of distinguished scholars management is implicated in a process which from around the world which analyses globalization, and reconfigures not only water supply regulation and calls for new types of economic governance. See Petrella urban waterscapes, but also the entitlements of (2001). water supply users, and the practice of deliberative 9 The UN's Committee on Economic, Social and Cultural democracy at the level of lived, daily experience. Rightsargued in its general comment (released26 November 340 Urbanization and water privatization in the South

2002) that the rightto water was implied in Covenantarticles Bauer C J 1998 Slippery property rights: multiple water uses 11 and 12, and that states party to the Covenant of Eco- and the neoliberal model in Chile, 1981-1995 in Arrojo P nomic, Social and CulturalRights (which entered into force and Martinez J eds El Agua a debate desde la Universidad: in 1976) have the duty to realize, without discrimination, hacia una nueva cultura del agua (Debating water from the the right to water (CESCR2002). The Covenant has been persepective of the University: towards a new culture of signed and ratifiedby 145 states, and signed but not ratified water) Fundaci6 Fernando el Cat6lico-CSIC, Zaragoza by seven states (includingthe United States)(UNHCR 2002). 607-48 10 The Constitutionof the Republic of South Africa guarantees Black J 1997 Oxford dictionary of economics Oxford Univer- the right of citizens of access to 'sufficientwater' (Act 108 sity Press,Oxford of 1996, section 7(2)). Blokland M, Braadbaart O and Schwartz K 1999 Private 11 Malin Falkenmark,a Swedish hydrologist,first proposed the business, public owners: government shareholders in water concepts of 'water stress' and 'water scarcity' in the 1980s, enterprisesMinistry of Housing, Spatial Planning, and Envi- in order to provide a baseline from which to measure ronment, Nieuwegein, the Netherlands and Water Supply threatsto water security;these benchmarksare now widely and SanitationCollaborative Council, Geneva accepted. 'Waterscarcity' is reached when a region has less Bond P 1998 Privatisation,participation and protest in the than 1000 m3 of renewable fresh water available annually restructuringof municipal services Urban Forum8 31-75 per capita. 'Waterstress' occurs when a region has between BraadbaartO 2001 'Privatizingwater: the Jakartaconcession 1000 and 1667 m3 of annual water availability per person and the lits of contract' Paper presented at KITLVJubilee (see, for example, Falkenmark1986 1989 1990). Workshop 'Water as a Live-Giving and a Deadly Force', 12 Water is subject to multiple marketfailures - most import- Leiden, 14-16 June antly, naturalmonopoly and externalities. BritishWater 1998 Britishwater - annual report 1997/98 British 13 A merit good may be defined as a good or service 'whose Water, London consumption is believed to confer benefits on society as a CESCR2002 Committee on Economic, Social and Cultural whole greaterthan those reflected in consumers' own pref- Rights adopts general comment on right to water Press erences for them' (Black 1997, 298). release 26 November United Nations Committee on 14 Winpenny, a research fellow at the government's Depart- Economic, Social and CulturalRights, Geneva ment for InternationalDevelopment (DFID),has writtenone Cook P and KirkpatrickC 1988 Privatisationin less developed of the few books on this subject to date (Winpenny 1994). countriesSt Martin'sPress, New York See also Merrett(1997). Cowan S 1997 Competition in the water industry Oxford 15 The P7 (now P8) annual conference was convened for the Review of Economic Policy 13 83-92 first time in June 1997 by the Green Group in the European Crespo C 2002 'La guerra del agua: nuevos movimientos Parliament,as an alternativeSummit to the G7 (now G8). sociales y crisis de dispositivos del poder' (The water war: Representativesfrom the world's poorest countries attend new social movements and crisis in power geometries) the conferences, which focus on the structural causes of Working paper presented at the PRINWASSconference, and solutions to poverty. St Antony'sCollege May 16 'Corporate'is here defined as an organization recognized DFID 1998 Better water services in developing countries: in law as a 'person', characterized by limited liability and public-private partnership - the way ahead Department bureaucratic hierarchy. In addition to private companies, for InternationalDevelopment, London many governmentsassume corporateforms (e.g. 'municipal Falkenmark M 1986 Fresh water - time for a modified corporations'). approach Ambio 15 192-200 17 See the Comisi6n para la Gesti6n Integral del Agua en FalkenmarkM 1989 The massive water scarcitynow threatening Bolivia (CGIAB) site (http://www.aguabolivia.org) for a Africa:why isn't it being addressed?Ambio 18 112-18 discussion of Bolivia's 'Ley de Exportaci6nde Aguas'. FalkenmarkM 1990 Global water issues confrontinghumanity Journalof Peace Research27 177-90 R 1997 Private sector participationin the water and References Franceys sanitation sector DFID Water Resources occasional papers Bakker K forthcoming An uncooperative commodity: privatiz- Waterand EngineeringDevelopment Centre and the Department ing water in England and Wales Oxford University Press, for InternationalDevelopment, Loughboroughand London Oxford Franceys R 2000 Water and public private partnershipsGlobal BakkerK and Hemson D 2000 Privatisingwater: hydropolitics WaterReport 31 March 16-17 in the new South Africa South AfricanJournal of Geography Giansante C, Babiano L and del Moral Ituarte L 2000 Evolution 82 3-12 des modalitds d'allocation des resources en eau en Espagne Barlow M and Clarke T 2002 Blue gold: the fight to stop (The evolution of modes of allocation of water resources in the corporate theft of the world's water The New Press, Europe)Revue d'EconomieMfridionale (REM)no. 2 New York Gilbert A and Gugler 1 1992 Cities, poverty and development: Batley R 1996 Public-privaterelationships and performancein urbanisationin the Third World 2nd edn Oxford University service provision UrbanStudies 33 723-51 Press, Oxford Urbanization and water privatization in the South 341

Goubert J P 1986 The conquest of water Polity Press, London identidades locales) Conference proceedings Colegio de Grusky S 2001 Privatizationtidal wave: IMF/Bankwater poli- Michoacin, M6xico, June cies and the price paid by the poor MultinationalMonitor 22 Petrella R 2001 The water manifesto: arguments for a world Gugler J 1997 Overurbanisationreconsidered in Gugler J ed water contractZed Books, London Cities in the developing world: issues, theory and policy Rivera D 1996 Privatesector participationin water supply and Oxford UniversityPress, Oxford 114-23 sanitation:lessons from six developing countries World Bank Haarmayer D and Mody A 1998 Worldwide water privatisa- Directions in Development, Washington tion: managing risks in water and sanitation FinancialTimes Roger N 1999 Recent trends in private participationin infra- Business Ltd,London structurePrivate Sector ViewpointWorld Bank, Washington, Haarmayer D and Mody A 1997 Private capital in water and September sanitationFinance and Development March 34-7 Rondinelli D A and Cheema G S eds 1988 Urban services Harvey D 1995 Globalization in question RethinkingMarxism in developing countries: public and private roles in urban 8 1-17 development MacMillan,London Hassan J 1998 A history of water in modern England and SawkinsJ 1995 Yardstickcompetition in the Englishand Welsh Wales ManchesterUniversity Press, Manchester water industry:fiction or reality?Utilities Policy 5 27-36 IFG2002 Alternativesto economic globalization:a better world Serra P 2000 Subsidies in Chilean public utilities World Bank is possible A reportof the InternationalForum on Globaliza- working papers in Infrastructure,Telecoms, Power, Water no. tion Berrett-KohlerPublishers, San Fransisco 2445 World BankGroup, Washington Inglese J L 2002 Presidentof AIDIS personal communication, Shirley M and Walsh P 2001 Public versus privateownership: Buenos AiresJanuary the current state of the debate Research paper World Bank, Johnstone N and Wood L 2001 Privatefirms and public water: Washington realising social and environmental objectives in developing Shiva V 2002 Water wars: privatization,pollution and profit countries EdwardElgar, London Pluto Press, London Kjellkn M and McGranahan G 1997 Urban water - towards Silva G, Tynan N and Yilmaz Y 1998 Private participation in health and sustainabilityBackground report for Comprehen- the water and sewerage sector - recent trends in Public sive Assessment of the FreshwaterResources of the World policy for the private sector Note 147 World Bank Group, Stockholm EnvironmentInstitute, Stockholm Washington 1-8 La Porte T R 1994 Large technical systems, institutional Solley W, Pierce R and PerlmanH 1998 Estimateduse of water surprises, and challenges to political legitimacy Technology in the United States in 1995 United States Geological Survey in Society 16 269-88 Circular1200 US Departmentof the Interior,Washington Laurie N and Marvin S 1999 Globalisation, neoliberalism, Swyngedouw E 1997 Power, nature,and the city. The conquest and negotiateddevelopment in the Andes: water projectsand of water and the politicalecology of urbanisationin Guayaquil, regional identity in Cochabamba, Bolivia Environment Ecuador:1880-1990 Environmentand PlanningA 29 311-32 and PlanningA 31 1401-15 Twort A C 1963 A textbook of water supply EdwardArnold, Lee T 1995 Financinginvestments in water supply and sanitation London NaturalResources Forum 19 275-83 UNHCR 2002 Status of ratificationsfor the principal interna- Lee T 1996 Alternativesfor privateparticipation in the provision tional human rights treaties Office of the United Nations of water services NaturalResources Forum 20 333-41 High Commissionerfor Human Rights,Geneva Lindh G and Gilbrich W 1996 Water in an urbanizingworld Van der Merwe S W 1995 An overview of water supply Nature and Resources32 31-4 managementin South AfricaAqua 44 151-60 Lovei L and Whittington D 1993 Rent-extractingbehaviour by WHO 1990 The international drinking water supply and multiple agents in the provision of municipal water supply: sanitation decade: review of decade progress World Health a study of Jakarta,Indonesia Water Resources Research 29 Organization,Geneva 1965-75 WHO 2000 Global water supply and sanitation assessment Marvin S and Laurie N 1999 An emerginglogic of urbanwater World Health Organizationand UNICEF,Geneva management,Cochabamba, Bolivia Urban Studies 36 341-57 Winpenny J 1994 Managing water as an economic resource McGuinness T and Thomas D 1997 The diversification Routledge, London strategies of the privatised WaSCs in England and Wales: World Bank 1994 Infrastructure and development: world a resource-basedview Utility Policy 6 325-39 development report, 1994 Oxford UniversityPress, Oxford Merrett S 1997 Introduction to the economics of water World Bank 1997a The state in a changing world: world resources:an internationalperspective UCLPress, London development report, 1997 Oxford UniversityPress, Oxford Nickson A 1997 The public-private mix in urban water supply World Bank 1997b Advancing sustainable development: the InternationalReview of AdministrativeSciences 63 165-86 World Bank and Agenda 21 World Bank,Washington Orellana R H forthcoming Politicas sobre recursos naturals World Bank 2001 PPIdatabase personal communicationJune y marco instituci6nal centralista en Bolivia: dos casos - World Bank/IFC1991 Office memorandum:privatization of the biodiversidady agues in Decentralization,municipalization, water sector in the UK and FranceWorld Bank/International and local identities (Descentralizaci6n, municipalizaci6n, Finance Corporation,Washington