You get what you can pay for:

Cost recovery and the crisis of service delivery in South Africa

David McDonald

paper presented at the Harold Wolpe Lecture Series

University of KwaZulu-Natal, Durban, 28 August 2002



(Please note that the fully annotated version of this paper is in the Centre for Civil Society online library)

Cost recovery refers to the practice of charging consumers the full (or near full) costs of providing services such as water and electricity. In direct contrast to the long-standing practice of subsidizing these services, where the state absorbs some or all of the costs of provision, service users around the world are increasingly expected to pay for the full costs of service delivery themselves.

This paper lays out the theory and practice of cost recovery in South Africa as it applies to basic municipal services such as water, electricity, sanitation and waste management. It draws on international literature and practice in this regard – particularly as it has been articulated by the World Bank and its partners – and illustrates how and why these policies have been introduced in post-apartheid South Africa.

The paper was originally published as the lead chapter in a book on cost recovery in South Africa (see McDonald and Pape 2002) and was written to provide a conceptual overview of what cost recovery means in practical and theoretical terms in advance of a series of empirical case studies in the rest of the book. The paper begins with an overview of what cost recovery means in practice and then reviews the fiscal, moral, environmental and commercial arguments used to justify its implementation. The paper concludes with an overview of the problems associated with cost recovery in the South African context, particularly as it relates to low-income households. As dry and mundane as it sounds at first blush, an examination of cost recovery takes us to the very heart of the neoliberal paradigm, with its focus on balanced budgets, fiscal restraint, market discipline, and privatization.

The paper is also intended as a preparatory discussion to the concluding section of the book that deals with “alternatives”. For it is only when we fully understand the models within which we operate that we can hope to develop different options.


What is cost recovery?

The concept of cost recovery is a simple one: the recovery of all, or most, of the costs associated with providing a particular service by a service provider. For publicly-owned service providers this may or may not include a surplus above and beyond the costs of production, while for private sector providers it necessarily includes a surplus (i.e. profit). In either case, the objective is to recoup the full costs of production.

Determining what to charge consumers is the difficult part. The devil, as the saying goes, is in the details, and it is to the pricing of services that we must first turn to better understand the basis of cost recovery.

For those services that can be accurately measured in volumetric terms (e.g. water, electricity, water-borne sewerage) cost recovery is achieved by charging end-users the (full) short-run marginal costs of production plus a portion of long-term operating and maintenance costs. To illustrate, if a person wanted to have electricity provided to their home, they would be expected to pay the costs of connecting the household to the electricity grid, a portion of the amortized operating and maintenance costs of the bulk infrastructure required to generate and distribute electricity, and a volumetric rate for the marginal cost of every kilowatt hour of electricity consumed.

There are several different ways of calculating these costs (Dinar 2000) but most models incorporate a downward sloping marginal cost curve where, due to economies of scale, those who consume more of a service are charged less per unit of consumption than those who consume less. In practice this has meant that poor households are penalized on a per unit basis because they consume less while wealthy households and industry benefit from the economies of scale.

In response to these equity concerns, progressive block tariffs have been introduced in many countries (including South Africa) in an effort to make the initial levels of consumption (or “blocks”) more affordable – or even free – while charging increasingly higher prices as consumption levels rise. This rising tariff has the added potential benefit of curbing consumption at the top end, thereby introducing conservation incentives.

Block tariffs are not inconsistent with cost recovery, however. The difference with more orthodox pricing models is that block tariffs charge higher than marginal cost prices at upper levels of consumption in order to make up for lower than marginal cost prices at low levels of consumption, effectively cross-subsidizing the poor. Most importantly, they also provide individual consumers with a certain level of subsidized consumption.

It is important to highlight, however, that not all services can be measured and priced on a volumetric basis. With services such as refuse collection and dry sewerage (as well as non-metered water or electricity) there is no way of accurately and easily measuring what an individual household has consumed. In these cases cost recovery models follow a flat rate that covers the average fixed and variable costs of the service. This can be done through a separate flat rate charge or can be included in a general rates account. Equity concerns can be dealt with through the application of differential rates, either along household income lines (i.e. means testing) or through some form of property valuation (i.e. the higher the value of your home, the more you pay, regardless of your level of consumption).

Whether volumetric or flat, all cost recovery models depend on “ring-fencing” – i.e. the isolation of costs and revenues associated with a given service and the removal of subsidies in or out of that sector. Ring-fencing means that resources – be they human or capital – cannot be shared between different service sectors unless they are paid for on a cost recovery basis to the unit that provided them (e.g. a water department would pay the accounting department of a municipality for the costs of keeping its books). The intent is to ensure that a service provider knows all its fixed and variable costs and is therefore able to apply (marginal) cost pricing to its consumers.

Managerially, ring-fenced units are controlled by officials who operate independently of all other service sectors and at arm’s length from elected authorities. Politicians generally retain the right to set standards and service delivery goals for a service unit, as well as monitor and evaluate their activities, but the daily management and long-term planning of the unit – including decisions about cost recovery – are done by the ring-fenced management team, whose only concern is the management of their own sector.

In theory, therefore, all of the costs of water, electricity, refuse and sanitation can be isolated and applied to end-users (with varying degrees of equity considerations). In reality, however, the actual costs of service production are seldom known: costs are complex and difficult to measure; they are constantly changing due to the lumpy nature of infrastructure investments; there are inevitably joint costs that are difficult to apportion; accounting for externalities is constantly evolving, and so on (Renzetti 2000, 130). At best, cost recovery models are an approximation of real costs.

In the end, it is fuller cost recovery that agencies such as the World Bank are after – charging prices that are as close as possible to the marginal costs in the short-term and the average cost curve in the long-term, with the aim of eventually achieving full cost recovery. I therefore use the term “cost recovery” to refer not only to full cost recovery but also to the intermediary stages of fuller cost recovery.


How is Cost Recovery Enforced?

For any cost recovery policy to be effective a service provider must be able to regularly and accurately measure the consumption of a particular service by an individual household, and it must be able to collect the payments. For volumetric services such as water and electricity measurement is relatively easy with the use of (increasingly sophisticated) meters that measure the number of litres used and kilowatt hours consumed. Without meters it is virtually impossible to apply marginal cost pricing. For those services that are not measurable on a volumetric basis it is necessary to approximate average consumption and charge the average cost (with or without differential rates).

But the most accurate measurement and pricing systems in the world mean little if the service provider cannot collect the monies owed for services rendered. Effective administration is important here – including a good postal/payment system – but so too are the punitive measures/threats used to persuade and force consumers to pay their bills. The most common form of punishment is to cut off a service to a household (or merely threaten to do so). In the case of water and electricity this means disconnecting the household from the water and electricity mains. In most situations this is temporary – turning off of a switch or valve that can be switched back on after the bills have been paid – but in an increasing number of “delinquent” cases in South Africa it involves shutting off services for weeks or months and sometimes the permanent removal of infrastructure to prevent illegal reconnections. Other enforcement tools include legal action, the attachment of assets, and, most controversially in the case of South Africa, the eviction from one’s home for non-payment of services. Cut-offs are more difficult with services such as refuse collection and non-water borne sewerage, but there have been cases in South Africa of households being denied these services as a penalty for non-payment of water or electricity (see Ruiters, this volume).

Cut-offs and evictions are expensive and politically sensitive enforcement weapons, however, which is why service providers interested in cost recovery are moving towards the use of prepaid meters wherever possible. A prepaid meter is a device that not only measures the exact amount of a service that is consumed – allowing for marginal cost pricing – it also forces users to purchase the consumption in advance. ‘Units’ (be it litres of water or kilowatt hours of electricity) are purchased at a retail outlet and then entered into the prepaid meter with the use of an electronic “smart card” (the meters are usually located at the household but can sometimes be centrally controlled).

Prepaid meters are the ultimate cost recovery scheme: they collect money in advance (and earn interest for the service provider in the process); do not allow the consumer to go into default; and they require no overt punitive measures. There are cases where prepaid meters have been tampered with, and the system can be cheated, but service providers and meter manufacturers are actively working to develop more sophisticated meters and to sell the prepaid meter as “pro-poor” (by arguing that it allows low-income households to budget more effectively for services and to avoid crippling debts). As one manufacturer of prepaid electricity meters stated to me in an email on the subject: “Without controlling the meter it is not possible to control the [electricity] business”.

The enforcement of cost recovery therefore requires a measuring system that allows a service provider to allocate costs to individual end-users, a billing system that informs consumers of their payment obligations, and collection mechanisms which ensures the payment of bills.


Cost Recovery in South Africa

This next section demonstrates that cost recovery on basic municipal services is a policy of national and local governments in South Africa. This has not always been the case, however. It is only since the end of apartheid in the mid-1990s that full (or fuller) cost recovery has been an explicit and widespread policy objective in the country.

There were user fees, tariffs and general property rates for services under successive apartheid regimes, but for the most part these charges had little relevance to the actual marginal costs of providing them. This was due in part to the fact that it was virtually impossible to estimate the costs of a given municipal service because apartheid local governments were so fragmented (the City of Cape Town, for example, which is now a single “unicity”, was composed of 25 municipalities and 69 local authorities prior to 1996). More importantly, there was little interest on the part of the apartheid state (national and local) to pursue full cost recovery. Indeed, the opposite was true with heavy subsidies (both hidden and transparent) being provided for services such as water, electricity, sewerage and refuse collection. This was particularly true for white suburbs and industry which received per capita infrastructure investments on par with, or even higher than, most European and North American countries during the 1970s and 1980s (Ahmad 1995) and yet continued to pay extremely low rates for these services.

Even the black townships and “bantustans” received considerable subsidies for services (although much smaller in relative and absolute terms than that of white areas). Part of this subsidization was direct, in the form of infrastructure developments and public housing in the 1960s and 1970s, but the bulk of it was indirect and in fact extracted unwittingly from the apartheid state in the form of payment boycotts by township residents in the 1980s and early 1990s. The apartheid state continued to provide services to these areas in spite of the boycotts– albeit in poor and deteriorating form – in fear of the political fall-out from not doing so, resulting in a de facto subsidization of township services.

These service subsidizations were no doubt motivated in part by clientelist politics (e.g. to win votes in an election; to keep puppet regimes in power in the “homelands”) but they were also driven by a “statist” vision of service delivery. In direct contrast to the neoliberal view of cost recovery and privatization that dominates official service delivery discourse in South Africa today, the apartheid state – be it national or local – saw its role as one of providing and subsidizing the delivery of essential municipal services (albeit in a racially skewed manner).

This statist model came under attack as early as the mid-1970s with the formation of the Anglo American-financed Urban Foundation, however, and began to splinter in the 1980s with the retreat of the National Party from state housing. But it was only in the 1990s that the subsidization paradigm began to crumble in a significant way. In fact, it has really been under the post-1994 African National Congress – both nationally and at the municipal level – that the push for cost recovery on basic municipal services has been most clearly and vociferously articulated. This shift has been supported and encouraged by an increasingly neoliberal civil service as well as an ideologically reconstituted New National Party (NNP) and overtly neoliberal Democratic Party (DP) (Bond 1999, Marais 2000, McDonald and Smith 2002).

This ideological transformation is not total, however. There are still some bureaucrats, politicians, and other members of the ruling elite who believe that basic municipal services should be delivered by the state and should be heavily subsidized. But the paradigmatic shift to privatization and cost recovery has been deep and profound. Virtually every political party in the country has expressed explicit support for cost recovery in policy documents – most notably the ANC and the Democratic Alliance (DA), which is a merger of the NNP and the DP – while implicit support from other quarters is evident in the silence that has met the introduction of aggressive cost recovery measures in the legislatures and councils of the country.

Most South African municipalities are still a long way from meeting these cost recovery goals – with payment rates in some areas as low 21% of billings – but the desire for full cost recovery is clear. In the meantime it is fuller cost recovery that the South African government is after, with “cost reflexive” pricing being the preferred nomenclature.

Take, for example, the directive in the Municipal Systems Act (RSA 2000a) – the omnibus legislation that deals with municipal services throughout the country – that service delivery should be as “cost reflexive” as possible (s74.2.d). The same applies to policies for specific services such as electricity, water and sanitation. In the Draft White Paper on Energy Policy (RSA 1998b, 7) it is stated that “Government policy is to…encourage energy prices to be as cost-reflective as possible.” In the White Paper on Water and Sanitation (RSA 1994, 19) it is argued that "government may subsidise the cost of construction of basic minimum services but not the operating, maintenance or replacement costs." The subsequent National Sanitation Policy White Paper (RSA 1996, 4) stated that: "Sanitation systems must be sustainable. This means…payment by the user is essential to ensure this."

Some policy documents make it clear that full cost recovery is the objective. The White Paper on Water Policy (RSA 1997, 4) proposed that in order to “promote the efficient use of water, the policy will be to charge users for the full financial costs of providing access to water, including infrastructure development and catchment management activities.”

It must be noted, however, that in each of these policy documents attention is paid to questions of equity in the form of indigency clauses, progressive block tariffs and, most recently, “free services” for an initial block of consumption. After stating that government will “charge users for the full financial costs of providing access to water”, for example, the White Paper on Water Policy (RSA 1997, 4) goes on to state that in order to “promote equitable access to water for basic needs, provision will also be made for some or all of these charges to be waived.”

The National Water Act (RSA 1998, Chapt. 5, Part 1) takes this equity issue further:

The Minister may from time to time, after public consultation, establish a pricing strategy which may differentiate among geographical areas, categories of water users or individual water users. The achievement of social equity is one of the considerations in setting differentiated charges. Water use charges are to be used to fund the direct and related costs of water resource management, development and use, and may also be used to achieve an equitable and efficient allocation of water.

Similarly, the Municipal Systems Act (RSA 2000a, s97.1.c) states that tariffs for municipal services can be differentiated based on indigency (i.e. poverty).

In other words, the cost recovery policy in South Africa has explicit equity considerations and distinguishes itself in this respect from more orthodox cost recovery models based on simplistic downward sloping marginal cost curves.

But as noted above, progressive block tariffs, indigent policies, and even free blocks of services, are not necessarily inconsistent with full cost recovery. The difference is where price points are placed vis-à-vis levels of consumption and at what point on the consumption scale consumers are expected to pay towards the full costs of service delivery. And as we shall see below, the manner in which block tariffs, free services and indigency policies have been introduced in South Africa have only marginally protected the poor while disproportionately benefiting the rich, and are very much constrained by being part of a broader cost recovery model within a neoliberal state (i.e. they are there to alleviate some of the hardest edges of cost recovery, not to replace it).

Another sign of the move to cost recovery in South Africa is the rapid and extensive introduction of meters – particularly prepaid meters – for volumetric services. Even communal water stands are being metered using prepaid systems in many municipalities, while in some areas communal taps are being removed and replaced with yard taps or in-house connections (for significant connection fees of up to several thousand Rands).

But the most visible feature of the move to cost recovery in South Africa comes in the form of punitive measures for non-payment of service bills. Service cut-offs, attachment of assets, and evictions from households are now common throughout the country, receiving considerable media attention. Research undertaken by the Municipal Services Project (McDonald 2002) suggests that as many as 10 million South Africans have had their water cut-off and 10 million have had their electricity cut off since the end of apartheid, with some two million people having been evicted from their homes for non-payment of service bills. Legislatively, The Municipal Systems Act grants local authorities this power, most notably the right to “seize property” for non-payment of services (s104.1.f).


The Rationale for Cost Recovery

Having established what cost recovery is in practical terms, and the fact that it is a formal policy of the South African government, we now turn to the question of ‘why’: what is the rationale behind the introduction of cost recovery on basic municipal services. The following sections provide a theoretical overview of the fiscal, moral, environmental and commercial arguments used by those who support cost recovery measures, and highlights how these arguments have been adopted in the South African context.


Fiscal arguments

The single most important reason given for cost recovery is the need to ‘balance the books’. Cost recovery, as the World Bank (1998, 44) is wont to say, is “a matter of good public fiscal practice”, allowing governments to reduce tax burdens and thereby attract and retain human and financial capital. Cost recovery in lower-income areas, it is argued, reduces the need for cross-subsidization from industry and higher-income households, making a country or a municipality a more financially attractive place to locate. These competitive pressures are often most explicit at the national level but are becoming increasingly common at the municipal level as cities and towns compete with each other for investment and struggle to deal with the downloading of responsibilities and cutbacks in inter-governmental transfers.

It is also argued that cost recovery is necessary to sustain services on a long-term basis. Without cost recovery, the argument goes, the state will not have the funds to invest in future service and infrastructure upgrades and extensions. Cost recovery is seen as “pro-poor” because it provides the fiscal basis for further service improvements and expansion: “When a public sector utility does not recover the costs of providing a service, it is often unable to extend the system – leaving poorer, marginal areas unconnected to the grid” (Brook and Locussol 2002, 37).

The South African government adopts the same basic lines of argument, stating in the Water Supply and Sanitation Policy White Paper (RSA 1994, 23) that if government does not recover operating and maintenance costs there will be a "reduction in finances available for the development of basic services for those citizens who have nothing. It is therefore not equitable for any community to expect not to have to pay for the recurring costs of their services. It is not the Government who is paying for their free services but the unserved."

These micro-economic policies are reinforced at the macro level with the fiscally-conservative Growth, Employment and Redistribution (GEAR) framework. Introduced by the ANC national government in 1996 without any consultation with its labour or civic allies, GEAR’s effect on cost recovery has been profound. First, it has resulted in significant decreases in intergovernmental transfers from national to local governments, resulting in an 85% decrease (in real terms) between 1991 and 1997 and further decreases of up to 55% between 1997 and 2000 (Finance and Fiscal Commission 1997; Unicity Commission 2000). Moreover, the amount of funding from central government is so low that it will take decades to address the backlogs. In fiscal 2000 total transfers to local governments across South Africa were only R3 billion. Projections of the capital costs required to address service backlogs, meanwhile, are in the order of R45 to R89 billion (depending on the level of services provided) with government-sponsored operating costs adding many billions more (RSA 1995, 2000c).

National government has also put caps on rates increases that local governments are able to levy on (wealthy) property owners. The Draft Local Government Property Rates Bill (RSA 2000b, chapter 2, sections 4-5), for example, states that local governments cannot apply taxes at the local level which threaten its own tax-reducing, fiscally conservative strategy, as evidenced by the following quotations:

A municipality may not…exercise its power to levy rates on property in a way that would materially and unreasonably prejudice national economic policies, economic activities across its boundaries, or the national mobility of goods, services, capital or labour.

The Minister, with the concurrence of the Minister of Finance, may by notice in the Gazette set a limit on the amount of the rate that municipalities may levy on property; or the percentage by which a rate on property may be increased annually.

With approximately 90% of all local government revenues being generated locally (of which approximately 25% come from property rates), these caps mean that local governments are unable to significantly increase their own revenue pools through progressive taxation. Little wonder, then, that local authorities have begun to push for fuller cost recovery as a way to finance and expand service delivery.

Competition for investment is a critical factor here as well (both within South Africa and internationally). As South Africa and its larger cities vie for increasingly mobile and fickle flows of private capital, municipal governments are under pressure to reduce tax and tariff rates in order to make it cheaper for firms to operate. As a result, there are pressures to minimize (if not eliminate and even reverse) cross subsidization measures in commercial, industrial, and high-income areas to assist with service delivery to the poor.

Consider the following. The Durban Chamber of Commerce and Industry “has in recent years expressed the concern of its members regarding the increasing cost of doing business in the Durban Unicity area. The cost of water is one of the major components of the total infrastructural cost, over which our members have no control….The prime concern of the Chamber is ensuring that the trading environment in the Unicity area contributes to the national and international competitiveness of its members in both commerce and industry”. The Chamber goes on to complain that the “expensive rural water schemes” in the newly-formed Unicity have “exacerbated the financial implications of possible solutions [i.e. put pressure on business to help pay for the costs of these water extensions]” and says that they will work to maintain “acceptable” bulk water tariffs (Anon. 2002).

Another revealing quote comes from the former Director General of the Department of Constitutional Development (the person in charge of all infrastructure development in the country): "If we increase the price of electricity to users like Alusaf [a major aluminium exporter], their products will become uncompetitive”.

In other words, if a country/city wants to be internationally competitive in terms of attracting private capital it must reduce subsidies (at least for the poor) and boost its cost recovery efforts.


Moral arguments

Another set of arguments used to justify cost recovery are moral in nature. The first of these revolve around liberal notions of rights and responsibilities. If people have the “right” to a service like water then they also have the “responsibility” to pay for it.

The South African Constitution and Bill of Rights are classic expressions of this thesis. “Everyone has the right”, for example, “to have access to sufficient food and water” (Bill of Rights, Section 27 (1)b). All South Africans would also appear to have the right to services that protect their basic health and well-being such as refuse removal and sanitation, as captured in the Bill of Rights environment clause (Section 24): “Everyone has the right to an environment that is not harmful to their health or well-being; and to have the environment protected, for the benefit of present and future generations, through reasonable legislative and other measures that prevent pollution and ecological degradation (for more on this point see Glazewski 2002).

The right to electricity is more difficult to ascertain but the Department of Minerals and Energy’s Draft White Paper on the Energy Policy of the Republic of South Africa (RSA 1998b, 3) does argue that “Energy should…be available to all citizens at an affordable cost”.

But these rights are met with obligations. The Department of Water Affairs and Forestry, for example, has taken out half-page newspaper advertisements entitled “Knowing your water and sanitation rights and obligations” in which it is clearly stated that one obligation is “paying your bills for services rendered”. According to the Municipal Systems Act (RSA 2000a, s5.2.6), this obligation applies across the board, with residents having the “duty” to pay for all of their municipal services.

Considerable efforts have gone into enforcing this message of civic responsibility in South Africa and tying it to a broader notion of civic and personal development. These goals are perhaps best exemplified by the Masakhane campaign (“let’s build together”). Introduced with much fanfare in 1995 by then-president Nelson Mandela, the campaign has focussed largely on convincing (low-income) consumers to pay their bills for water, electricity and other municipal services. But the ANC has also been at pains to point out that Masakhane is about more than bill payments. It is seen as a broad political vision that makes service payments part of a larger societal transformation: “The true essence of Masakhane [is] mass involvement in the transformation process” (ANC 1997).

A related argument is to be found in the burgeoning “willingness to pay” literature. The rationale here is that most people – low-income households included – accept their civic responsibility to pay the full costs of service delivery, and are happy to do so as long as the services are reliable, affordable and of good quality (see, for example, Jimenez 1987; Whittington, Briscoe and Mu 1990; Whittington, Lauria and Mu 1991; Alberini and Krupnick 2000).

Finally, it is argued that only by paying the full cost of a good or service can one appreciate its true “value”. Receiving a service for free, or having it heavily subsidized, distorts not only its exchange value but its use value as well; the very essence of the thing being consumed. According to the World Bank (1998, 44), only “a fee reflecting the costs will encourage users to correctly value the service they receive”. Charging a fee will “help reverse the ‘entitlement mentality’ that has been the historical result of subsidizing public services.”

The notion that a service such as water could have a use value without an exchange value – i.e. a right without a (financial) responsibility – is an alien one to the commodity-oriented cost recovery literature. This final point takes on a proselytizing tone in South Africa, with private capital and (apartheid-era) municipal bureaucrats keen to inculcate market values into township dwellers and rural Africans through tough cost recovery measures that will educate low-income households, like naïve children, about the “correct value” of municipal services.


Environmental arguments

These moral arguments are extended to the environmental arena. Subsidization, it is argued, promotes wasteful consumption of environmentally-sensitive services such as water, electricity and refuse collection because the “correct value” is not reflected in the price, meaning that there is little financial incentive to limit consumption. In other words, subsidies promote waste while cost recovery promotes conservation.

The World Bank (1994, 83) has the following to say about the environmental benefits of cost recovery in the area of electricity:

Efforts to mitigate environmental impacts through consumer investments in energy saving are hampered by low consumer prices and subsidies. On average, developing countries use 20 percent more electricity than they would if users paid the incremental cost of supply. Once economic pricing is established, governments are able to promote the use of more energy-efficient technologies.

The argument that prices are the best - indeed the only - way to shape human behaviour is central to all (neo)liberal tropes, from Adam Smith onwards. Without price incentives and market institutions humans behave in self-maximizing and destructive ways (the “tragedy of the commons” thesis). Proper pricing, on the other hand, creates a moral and economic framework for environmental sustainability by appealing to our own self interest.

The World Bank (1994, 81) argues that these price incentives should be structured in the form of rising block tariffs (in order to curb over-use by the rich). With downward sloping cost curves for many services this is in fact the only way that conservation can be achieved through pricing mechanisms. In reality, however, the World Bank (and the South African government) do not always promote block tariffs – or, more accurately, they do not promote sufficiently progressive block tariffs with prices that would really make a difference to consumption at the upper end – and thereby undermine the conservation potential of this economic framework. This is especially true of commercial agriculture and industry in South Africa which have been amongst the most inefficient users of water and electricity, and flagrant producers of waste, in the world.


Commercialization arguments

Last but not least is the argument that cost recovery promotes efficiency, accountability and transparency by providing easy to understand performance indicators: a financial surplus means success; a deficit means failure. Subsidies, by contrast, obscure the bottom line, making it difficult to evaluate service performance in a given sector and contributing to bureaucratic sloth, mismanagement and fraud.

This introduction of business principles is a key part of the neoliberal transformation of municipal services. Broadly defined as the “commercialization” of services, these principles apply as equally to publicly-owned, ring-fenced service units (i.e. corporatization) as they do to public-private partnerships, outright privatization, and everything in between.

The stated rationale here is that finances are the only true and reliable indicator of service performance (just as they are the only true indicator of a service’s “value”). Cost recovery, it is argued, forces managers to think continuously about the bottom line, engenders creativity, forces transparency and provides an incentive to constantly improve service delivery through performance-based salaries – a system where managers are evaluated and remunerated based primarily on financial results.

But cost recovery must also been seen as a crucial step in the privatization of municipal services. For without full (or close to full) cost recovery in the public sector there is little chance of furthering the cause of privatization. No private firm is interested in purchasing a public utility that only collects 50% of its costs in revenues (unless they can obtain subsidies from the state to make up the difference, which is not uncommon).

Speaking to a UNDP/World Bank workshop on financing water delivery, Mike Muller (1999, 4), the Director General of the Department of Water Affairs and Forestry (DWAF) in South Africa, had the following to say about the relationship between cost recovery and privatization:

I must be blunt. There are many who believe and/or hope that the private sector may make money available for basic levels of service in poor communities. But we must remember, the private sector cannot (and should not) be bankrupted; it is there to make a profit. So we should not imagine that private sector involvement is somehow going to make it possible for the poor to get what they cannot afford to pay for - at least not directly. In other words, the private sector will not (and “should not”) provide services to the poor at below cost. Cost recovery is paramount and services will only be provided by the private sector to those who can pay, or will be provided at a level that people can afford (e.g. communal water taps, pit latrines).

Having said that, it would be incorrect to conflate cost recovery with privatization. “Cost reflexive” service providers can remain public and still satisfy the fiscal restraint objectives of neoliberalism. It could even be argued in the current South African context that public sector service managers are more aggressive than their private sector counterparts when it comes to cost recovery given their desire to ‘prove’ themselves as being business-minded (either because of the political and media pressure they are under to do so and/or because they are being evaluated and remunerated based on cost recovery measures).

The public sector also tends to service more low-income areas – due to “cherry-picking” in the private sector – and therefore have to deal with the more difficult cost recovery cases, making their cost recovery efforts more visible.


Problems With Cost Recovery in South Africa

I have alluded to a number of concerns with cost recovery thus far but turn now to a more explicit critique of how it has been applied in South Africa. The discussion is broken into two parts. The first describes a series of practical concerns with cost recovery and highlights the inequitable manner in which it has been applied. South African cost recovery policy and legislation makes a rhetorical commitment to equity, but when it comes to implementation it has been anything but. In highlighting these concerns I also anticipate some of the short- to medium-term “reformist alternatives” to cost recovery – quick and important ways of alleviating some of the most regressive aspects of cost recovery as it is currently practiced.

The final section looks at the larger theoretical question of commodification and the distorting effect that this has on our ability to develop systems of production and consumption of essential goods and services outside of the dictates of the market.


Historically Unfair

Perhaps the single most compelling argument against full cost recovery in South Africa is the fact that it was not practiced under apartheid. As noted earlier, white South Africans and industry benefited enormously in social and economic terms from heavily subsidized municipal services. Now that apartheid has officially ended, black South Africans are expected to pay their own way, with scarcely a mention of the fact that this has not always been state practice. The ahistorical nature of policy development and the hard-hitting tone of the Masakhane campaign reek of hypocrisy in this respect, with the biggest beneficiaries of the formerly-subsidized system - white rate payers and industry - being the most vociferous advocates of a user-pays system.


Continued Subsidies for the Rich

Inequality in cost recovery it is not just an historical concern, however. There are clear and significant examples of ongoing pricing biases in favour of suburban residents and industry. There are, for example, enormously different tariffs for electricity in South Africa, with rural (African) households paying on average 48c per kwh while suburbanites pay an average of only 23c. Township residents will often pay more for electricity than their suburban counterparts as well, despite being in the same city. Residents of Soweto, Johannesburg, for example, were paying about 30% more per kwh than residents in the nearby wealthy suburb of Sandton (Maj-Fiil 2001).

Industry benefits the most in this regard. To illustrate, average electricity prices for the manufacturing and mining sectors in South Africa are in the order of 12c per kwh, while deals are negotiated with large consumers (e.g. Alusaf) for prices as low as 3.5 cents per kwh (Fiil-Flynn 2001, 4). In fact, South African industry enjoys the lowest industrial charges for electricity in the world.

Part of the reason for these price differences is that capital costs in most suburban and industrial areas are sunk and fully amortized, while in township and (especially) rural African areas prices reflect the full capital costs of new infrastructure development and upgrading. In other words, suburbs and industry continue to benefit from the racially-skewed investments of apartheid due to a pricing structure that has largely written off their fixed costs. There are competitive reasons at play here as well, with the South African government keen to keep industry globally competitive, even if this means subsidizing their electricity at below cost.

There are also hidden subsidies for suburbs and industry. One of the most significant is the ongoing skewed nature of municipal spending and resources distribution. Recent research by the Municipal Services Project has shown enormous differences in resources available for service delivery in suburbs and industrial areas as compared to townships. In Cape Town, for example, discrepancies of five, ten and even one hundred-fold were not uncommon in the water and waste management sectors (McDonald and Smith 2002), with similar results in the waste management sector in Johannesburg (Barchiesi 2001). In effect, township residents are paying the same (or higher) per unit prices for a given service as suburban residents, while in practice receiving considerably lower resource allocations.


Voluntarism as a Form of Cost Recovery

Another form of inequality is found in the guise of voluntarism. Using the argument that low-income households/neighbourhoods are unable to afford the full costs of a particular service (or a particular level of service), municipalities will often ask local residents to contribute their labour on a volunteer basis to lower the costs of service delivery. Rather than paying with their wallets, these volunteers pay with their blood, sweat and tears, picking refuse off the street, cleaning sewers, digging ditches, and so on. Not only are there hidden health and social costs associated with this kind of work – with the highest price being paid by women as the ones most likely to “volunteer” – there are also race and class dimensions to this form of cost recovery which make it little more than neo-apartheid in nature. Take, for example, the following quote from a rural dweller:

The RDP [Reconstruction and Development Programme established by the ANC in 1994] is ridiculing our mothers. Our mothers are made to dig trenches. It is called employment. Whereby you walk right around this South Africa and you never find a white woman digging a trench. The dignity of our mothers is taken because they have to dig trenches, while they have to feed their babies, cook for their loved ones (cited in Budlender 1998, 21).


Progressive Block Tariffs?

Despite the lip service paid to rising block tariffs for water and electricity, few municipalities in South Africa have introduced them and enforced them in a meaningful way. Most block tariffs rise steeply in the first one or two blocks after the “lifeline” or free block – penalizing those at the lower end of consumption – while tapering off or even decreasing at the top end. Some municipality’s tariff structures are regressive right from the beginning, with prices decreasing as consumption increases.

But even where tariff structures are rising they are unlikely to make any significant difference to the consumption patterns of well-to-do suburbanites as they are currently structured. In Cape Town, for example, which has one of the most progressive water tariff structures in the country, a household will pay only R212.40 for consuming 60kl litres of water per month – an enormous amount of water to be sure (used for watering gardens, filling pools and washing cars), but hardly a budget-breaking expense for the typical middle- to upper-income suburban family with monthly household earnings of R6000 or more. Meanwhile, a household in the townships consuming 20kl of water per month would be paying R36.40 – a considerably lower payment in absolute terms but one that is likely to be much higher in proportional terms given typical household incomes in the townships of R500 per month or less.

In fact, it is not uncommon for low-income households to spend 25-40% of their incomes on basic municipal services (McDonald 2002).

Block tariffs in South Africa must be considerably more progressive if they are to play an effective role in cross-subsidization and conservation, and they will need to be more consistent across (and within) municipalities in order to avoid capital flight to places where tariffs for industry and high-income households are low.


Free Services?

A related problem is the issue of “free services”. Developed initially by the national office of the ANC in the lead-up to local government elections in December 2000, and subsequently adopted by the Democratic Alliance as part of their election campaign in the same year, the free services policy is a variation on block tariffs with the initial block of consumption in water and electricity being provided at no cost (six kilolitres of water per household per month and 50 kwh of electricity per household per month).

Although potentially progressive, and an important step forward in the cost recovery debates in South Africa, free services in practice would appear to have made little difference to the lives of the urban and rural poor since being introduced in mid-2001. First, there is the problem of the quantity of free services being offered. With respect to electricity, the 50 free kwh per household per month being offered by ANC-controlled municipalities (and only 20kwh by DA-controlled municipalities) will provide some financial relief, but this amount of electricity will only run a light bulb and a few small appliances for a month (one kilowatt hour will light a 100 watt light bulb for ten hours). Moreover, 50% of rural families are not on the electricity grid, meaning that, for some time to come, millions of low-income households that will not benefit at all.

The promise of 6kl of water per household per month also offers little financial respite due to the fact that many low-income households use considerably more than six kilolitres due to relatively high average numbers of occupants per household and also because of old and leaky apartheid-era infrastructure. Rapid tariff increases after this free block can mean that poor families end up paying more, not less, for water than they did under old tariff structures, while those accustomed to paying a “flat rate” for services have seen dramatic price increases for both water and electricity (more than 400% for the cost of electricity in some cases in Soweto, despite a 15% decrease in the average real price of electricity since 1994 (Fiil-Flynn 2001)).

Another problem is that the 6kl figure is based on an average household of eight people and works out to 25 litres per person per day. This 25 litre figure is at the bottom end of the World Health Organization’s (WHO) recommended daily minimum – with estimates of 50 litres being more common in the health literature – and is well below the 50-60 litres per day called for in the ANC’s original Reconstruction and Development Programme (RDP) as a medium-term service delivery goal. To put this in perspective, the average bathtub takes 200 litres to fill while the average toilet uses 10 to 15 litres per flush — a situation made worse by the fact that water (and energy) saving devices have never been a serious part of service delivery strategies in South Africa.

The fact that many low-income households have more than eight people heightens the problem, as does the fact that for many people this water is only delivered to a communal, metered standpipe within 200 meters of the recipient’s household (as part of the “basic level” of services envisioned by the ANC). With violence and rape a serious problem in many low-income areas, these communal standpipes can be both inconvenient and unsafe, particularly after dark.

There is also a concern with the use of the household as a unit of measurement for free water and electricity due to its intrinsic bias against low-income families. To illustrate, a young couple with two incomes and no dependents living in a home in the suburbs receive the same amount of free water as a single, unemployed mother with seven dependents living in a run-down council house or shack in the townships. In many municipalities (e.g. Durban) households are not means-tested to see if they qualify for the free service (using the rationale that the administrative cost of these tests would outweigh the savings) with the result that some middle- and upper-income South African households are benefiting more from the provision of free lifeline services than poor households. This is not to suggest that individual means-tests should be used to determine which households should have access to free services – a potentially degrading and divisive procedure separating the very poor from the even poorer – but it does highlight the inherently inequitable feature of basing free services on a per-household basis.

Finally, there is the problem of delivery. Although free water and electricity were to have been implemented across the country on July 1, 2001, implementation delays have been widespread – particularly in rural areas – and there have been disputes over what level of government should cover the costs of free services. In the case of electricity, the roll out has been further hampered by unresolved negotiations between the parastatal Eskom and national government over the subsidization of the free 50 kwh, resulting in a lengthy delay for free electricity in Soweto and other township and rural areas.

Moreover, many households are not receiving free blocks of water and electricity because of they are in payment arrears, and there are widespread reports of continuing cut-offs of water and electricity despite the free services policy. The Department of Provincial and Local Government (DPLG 2002, 30-1) reported that more than 296,000 disconnections of electricity and 133,000 disconnections of water took place in the last quarter of 2001 – most of which would have been low-income households – and this at a time when “free services” promises had been in effect for several months. And these figures do not tell the whole story because the data are incomplete (not all municipalities reported) and there were no statistics from Eskom which had been cutting off as many as 20,000 households a month in Soweto in early 2001 (Fiil-Flynn 2001).


Narrow Accounting Methods

Even if block tariffs were much more progressive, municipal resources were more equitably distributed, and free services were more accessible, neoliberal cost recovery models (South Africa’s included) are fundamentally flawed by their narrow accounting methods. Only direct financial costs are included in the cost recovery analysis, leaving out many of the less tangible but equally important costs and benefits of service delivery such as gender equity, public health and safety, and the environment. As a result, pricing structures tend to overcharge the poor (for whom lower prices would generate considerable private and public good benefits) and undercharge the well-to-do (who have never paid the real social and environmental costs of their hedonistic consumption patterns in South Africa). To account for all of these social and environmental costs would dramatically alter the cost recovery equation.

One of the reasons this is not done is that these social and environmental costs are very difficult to quantify. What kind of a monetary value, for example, does one attach to a valley that has been flooded out by a new dam, or the loss of dignity associated with service cut-offs, or the psychological scars of a rape as a result of collecting fuel wood at night along dark paths? These are real costs, nonetheless, and must be taken into account.

Some costs are more easily measured, such as public health epidemics. One recent study has shown that the costs of dealing with all diarrhoea-related illnesses in South Africa in 1995 (much of which was a direct result of poor water and sanitation services but no doubt exacerbated by water cut offs) was in the order of R3.4 billion per year in direct medical costs and R26 billion per year in lost economic production – more than the total amount needed to provide water infrastructure to everyone in the country over a 10 year period (Moodley 2000).

And yet, even these relatively easy to measure costs are ignored in the drive for cost recovery in South Africa. One tragic example is the cholera outbreak that began in mid-2000. As Cottle and Deedat (2002) argue, it was the introduction of cost recovery efforts on water in rural KwaZulu-Natal that precipitated (or at least exacerbated) the cholera crisis by forcing households that could not afford the new connection fees and volumetric charges for water to use contaminated ponds and rivers for bathing and drinking. The outbreak has been the worst in South Africa’s history and has led to over 100,000 illnesses and close to 300 deaths. Ironically, the state is now spending tens of millions of Rands dealing with the epidemic, many times more than it was costing the provincial government to provide free (unmetered) communal tap water prior to the cost recovery measures.

Ring-fencing is partly to blame here as service managers focus on their own, narrow sectoral bottom lines, uninterested in and/or uninformed about the implications of their decisions on other sectors. There is a fundamental contradiction here, therefore, between the drive to ring-fence a service so it can better isolate its own micro costs, and the need to understand the braoder macro-economic costs and benefits of a particular service (e.g. public health).

Nor can one forget the fact that national and municipal governments are under increasing pressure to keep operating costs as low as possible for footloose international capital. To include the true costs of gender equity, environmental sustainability and spatial desegregation into water and electricity tariffs would undoubtedly put South Africa and its cities at a competitive disadvantage. Instead, government has made it clear in its 1998 White Paper on electricity that prices to industry must not be affected by the (true) costs and benefits of a household electrification programme: “Cross subsidies should have minimal impact on the price of electricity to consumers in the productive sectors of the economy”.


Harsh Measures

Another concern with the way that cost recovery is being implemented in South Africa is the harsh way in which it is being enforced. From letters threatening legal action, to cut-offs and evictions, there has been a considerable amount of intimidation and even overt violence associated with cost recovery in the country. In some cases private companies are hired to remove people’s infrastructure and furniture (the so-called “red ants” of Johannesburg’s townships) and there have been incidences of significant physical abuse and intimidation as people resist cut-offs and evictions (e.g. tear gas, rubber bullets and real bullets used against anti-eviction protestors in the township of Tafelsig, Cape Town, in September of 2001; the shooting at members of the Soweto Electricity Crisis Centre in Johannesburg in April 2002 as they protested against cut-offs in front of the mayor’s house. See also Desai for examples in Durban (2000)). Some families have come home at the end of the day to discover that their home has been auctioned for non-payment of services and their belongings are on the street (Deedat, Pape and Qotole 2001). In other cases, people have received bills for thousands of Rands for water or electricity that they did not use but ended up on their bills because of massive leaks in pipes, faulty meters, or incorrect meter reading, but had their services cut or were evicted from their homes anyways (see Fiil-Flynn 2001 for examples in Soweto).

There are more subtle forms of injustice and insult as well: pensioners waiting for hours on end in the heat and cold to pay their bills at under-staffed municipal offices and then getting rude service when they finally manage to speak to someone; a lack of flexibility when it comes to payment schedules despite being on (low) fixed incomes; being told that you should be used to living without water and can get by a little longer; and so on.

These are not merely isolated incidents. As noted earlier, cut-offs and evictions have been widespread in the country, with millions of people having experienced a water cut-off, an electricity cut-off, or an eviction from their home for non-payment of service bills. Not all of these incidents have been violent, but the emotional impact of having to go without water and/or electricity – sometimes several times a year, sometimes for months on end – is considerable and constitutes its own kind of structural violence. The fact that the main electricity provider – Eskom – reported a 37% increase in profits to R2.56 billion in 2001 simply adds insult to injury.

Most insidious of all, however, are the hidden injustices of prepaid meters. Far from being friends of the poor – as municipalities such as Cape Town insist they are, arguing that they help low-income households budget for services and avoid crippling arrears – prepaid meters merely conceal the extent of service cut-offs by having low-income households cut their own consumption at the point of purchase. Although it is too early to know the full impact that this new technology may have, if the experience in the UK with prepaid meters is anything to go by (Drakeford 1998) it is very likely that they will yield yet another layer of inequity in South Africa.

Finally, the primary targets of these harsh cost recovery efforts are low-income households, despite the fact that large commercial and industrial consumers are often the most delinquent, and the largest, payment defaulters. In the City of Cape Town, for example, businesses and industry make up just under a third of the R2.1 billion outstanding in service payments. In one instance, the clubs that run the Newlands cricket ground – the premier destination for “world class” cricket in the country – have been allowed to run a deficit of some R4 million in service arrears, and there has even been a proposal by the city to offer tax rebates for the stadium. Meanwhile, low-income families have been evicted from their homes in the city’s townships for arrears of only R1500 and some have received threats of eviction for arrears as low as R250!



Not only are these practices morally bankrupt (particularly when one considers the strong moral overtones of the cost recovery agenda), they are very possibly unconstitutional. Although the Bill of Rights does not provide for equity based on income (see the Introduction to this volume) it does provide the right to a “healthy and safe environment” and “access to adequate housing”, under which access to basic municipal services would seem to apply. Indeed, the landmark “Grootboom” case in the Constitutional Court in May 2000 has set a precedent for socio-economic, or second-generation, rights as enshrined in the Constitution (access to housing in this case) and may be used in future court challenges on service cut-offs and evictions.

One international indication of the potential for this constitutional/legislative change is found in the United Kingdom where water cut-offs to residential homes, schools and other essential public buildings has been outlawed since 1999 after a rash of cut-offs had taken place when water services were privatized. It has also been ruled in the UK that self-imposed cut-offs through the use of prepaid meters is illegal and it is the responsibility of the water supplier to ensure water provision even if the household is unable to pay for prepaid consumption (Drakeford 1998).


Blindly Ideological

Finally, it is worth noting that much of the cost recovery efforts to date in South Africa have been driven by a blind ideological faith in neoliberalism. There has been no thorough analysis of the costs and benefits of cost recovery, and certainly no dedicated effort to explore alternatives, aside from the politically-expedient decision to offer “free” water and electricity in the run-up to local elections in December 2000. Nor has there been any concerted effort to restructure and redistribute existing municipal resources which, as noted above, remain highly skewed along racial lines from the apartheid era and inefficiently deployed. Local governments have been virtually silent on the dramatic cutbacks in inter-governmental transfers as well, effectively accepting their fate of unfunded mandates, and have been pushing hard to collect the full costs of municipal services without challenging the state for a larger slice of the national budget. The DA in Cape Town did threaten to make this an issue in the lead up to the December 2000 local elections but have never followed up on it – no doubt a result of their own commitment to macro-economic restraint at the national level.

So too are municipalities – especially the large metropolitan ones – blindly committed to a notion of being globally competitive. Barely a report from any of the large municipalities is released without some reference to being “world class” and “a good place to invest” (for an analysis of Cape Town see McDonald and Smith (2002)).

But it is the lack of monitoring and evaluation of cost recovery that is perhaps most disturbing of all. This book represents the first concerted effort in South Africa to measure and evaluate the impacts of cost recovery and provides the only comprehensive set of quantitative and qualitative data available. The Department of Constitutional Development (now the Department of Provincial and Local Government) did run a tracking programme in the mid-1990s called “Project Viability” which included some questions for municipalities about the number of cut-offs they had undertaken each quarter, but reporting rates were as low as 30% of municipalities and the programme stopped after a few years. There was no effort by the Department to evaluate the qualitative impact of these cut-offs in terms of their effect on the lives of the urban and rural poor.

The DPLG has begun to collect this raw data once again – starting in mid-2001 – but has not made it available to the public through their website as they did with earlier Project Viability data. The results cited earlier – more than 296,000 electricity disconnections and 133,000 water disconnections in the last quarter of 2001 – are included in this book only because we managed to get access to the report via a researcher. The data are also incomplete, with only 88% of municipalities reporting, and there are inconsistencies in data presentation. In the end, it is not clear that cost recovery advocates in the government understand, or even know about, the implications of their disconnections policies.

Nevertheless, national and municipal bureaucrats and politicians forge ahead, accepting the mantra of balanced budgets, ring-fencing and commercialization, while listening intently to advisors from the World Bank and other neoliberal donor institutions and pay millions of Rands to consultants from the private sector to help them with their cost recovery plans.



Many of the concerns cited here can be mitigated to a greater or lesser extent by remedial policies and actions of local and national governments: more progressive tariff structures; a better distribution of existing municipal resources; more flexibility in payment schedules and/or writing off of arrears; more holistic accounting methods; larger quantities and a wider range of “free” services; new legislation banning water cut-offs to residential housing and public buildings; a dedicated commitment to collecting and analyzing data on the impacts of cost recovery. These and other policy measures can be enacted in South Africa relatively easily and could make a significant difference to the lives of millions of people in the short- to medium-term (see McDonald and Pape (2002), chapter nine, for a more detailed discussion).

But as difficult as it will be to enact these reformist policy measures in South Africa, it is the longer-term question of “de-commodification” that will prove to be much more intractable – i.e. removing “price” altogether as a determinant in the production and consumption of services like water and electricity. De-commodification is a call that is being heard with increasing frequency in South Africa and in the anti-neoliberal literature more generally (e.g. Barlow and Clarke 2002; IFG 2002). The rationale is that societies the world over have – and to some extent still do – determined resource distribution by means other than exchange value, making decisions based on notions of the commons, shared cultural values, etc. Although not unproblematic in themselves (e.g. gender relations can be unequal in “traditional” cultures) these non-commodified systems of production and distribution nonetheless provide powerful alternatives to the neoliberal cost recovery framework.

Are they feasible in South Africa? One immediate concern is what would be done with rich households and industry. Would these consumers also be given access to “free” water and electricity? Would their “shared cultural values” fit with those of low-income families? What would be the power relationships between multinational corporations and households when it came to determining distribution of water on a basis other than price? Clearly there are some very real concerns here that would require an enormous shift in institutional decision making, political culture, and so on; a shift that would take many years to prepare for and introduce.

Partial de-commodification is not going to resolve these tensions either – i.e. the removal of price from a few key services like water and electricity. The problem is that these services still function within a larger market economy – an economy that is constantly (re)inventing new products and needs and requiring ever-increasing volumes of supply of these goods and services in production and consumption (e.g. silicon chips, dishwashers), putting constant upward market-driven pressure on their use and allocation. The domestic use of water and electricity, therefore, cannot be separated out from the broader production and consumption cycles of capitalism and the role of prices in determining what gets produced for whom.

In other words, de-commodification is an all or nothing proposition. Either everything gets de-commodified, or nothing gets de-commodified (only tampered with around the edges through the use of subsidies, lifeline tariffs, etc.). De-commodification takes us to the heart of the market economy, built as it is on the transformation of “things” (with a “use value”) into products (with an “exchange value”) through the use of wage labour. To challenge this transformation process, and the impact it has on growth and the allocation of resources, is to challenge the very foundations of capitalism (Harvey 1982).

A massive proposition indeed, and one that takes us far from the seemingly technical challenges of a more equitable pricing system. These are essential considerations, nonetheless, as South Africa grapples with the question of what makes for a fair and sustainable system of producing and consuming basic municipal services.



I would like to thank the following people for their comments on an earlier draft of this paper: Peter McInnes, John Pape, John Williams, Patrick Bond, Laila Smith, Alex Loftus.



Alberini, Anna; Krupnick, Alan J . Cost-of-illness and willingness-to-pay estimates of the benefits of improved air quality: evidence from Taiwan, Land Economics v 76 no1 Feb 2000.

ANC 1997, “Supplement”, Mayibuye: The Journal of the ANC, October.

Ahmad, Junaid. "Funding the Metropolitan Areas of South Africa." Finance and Development, September, 1995

Anon. “Water Costs Increases”, Chamber Digest – Official Newsletter of the Durban Chamber of Commerce and Industry, Issue 3, February 16, 2002, p. 9

Arimah, Ben C, Willingness to pay for improved environmental sanitation in a Nigerian city, Journal of Environmental Management v 48 Oct 1996

Brook, Penelope J. and Alain Locussol (2002), “Easing tariff increases: Financing the transition to cost-covering water tariffs in Guinea”, in Contracting for public services: Output-based aid and its applications, Edited by Penelope J. Brook and Suzanne M. Smith, World Bank: Washington, D.C.

Barlow, Maude, and Tony Clarke, “Blue Gold: The Battle Against Corporate Theft of the World's Water”, Stoddart Press: Toronto, 2002.

Budlender, Debbie. 1998. The People’s Voices: National Speak Out on Poverty Hearings, March to June 1998. Johannesburg: Commission in Gender Equality, South African Human Rights Commission and South African NGO Coalition.

Deedat, Hameda, John Pape and Msokoli Qotole, 2001, Block Tariffs or Blocked Access? The Greater Hermanus Water Conservation Programme, Municipal Services Project Occasional Papers Series, No. 5, Cape Town.

Democratic Alliance, 2000. Local Government Manifesto – “For ALL The People”, December, mimeo.

Desai, Ashwin. The Poors of Chatsworth. Madiba Publishers: Durban, 2000

Dinar, Ariel (ed), 2001, The Political Economy of Water Pricing Reforms, Oxford University Press (for the World Bank), Washington, D.C.

DPLG [Department of Provincial and Local Government], (2002). Quarterly Monitoring of Municipal Finances and Related Activities, Summary of Questionnaires for Quarter Ended 31 December 2001, Pretoria, mimeo.

Drakeford, Mark. “Water Regulation and Pre-payment Meters”, Journal of Law and Society, Vol. 25, No. 4, December, 1998.

Eberhard, Anton and Clive van Horen. Poverty and Power: Energy and the South African State. Johannesburg: Pluto Press, 1995.

Fiil-Flyn, Maj. 2001. The Electricity Crisis in Soweto, Municipal Services Project Occasional Papers Series, No. 4, Cape Town.

Harvey, David. The Limits to Capital. Basil Blackwell: Oxford, 1982

Hui, Eddie Chi Man , Willingness to pay for better housing in Hong Kong: theory and evidence, Urban Studies v 36 no2 Feb 1999.

IFG [International Forum on Globalization], “A Better World is Possible: Alternatives to Economic Globalization – Report Summary”, IFG: San Fransisco, 2002.

Jiminez, Emmanuel. Pricing Policy in the Social Sectors: Cost Recovery for Education and Health in Developing Countries, Baltimore: Johns Hopkins University Press for the World Bank, 1987

McDonald, David A. 2002. “The Bell Tolls For Thee: Cost Recovery, Cut-offs, and the Affordability of Municipal Services in South Africa”, in David A. McDonald and John Pape (Eds.), The Crisis of Service Delivery in South Africa, HSRC Publishers: Pretoria and Zed Press: London.

McDonald, David A., and John Pape (Eds.). 2002. The Crisis of Service Delivery in South Africa, HSRC Publishers: Pretoria and Zed Press: London.

Moodley, Shomenthree, 2000, Investigating Total Economic Burden: South Africa’s Diarrhoeal Disease Burden in 1995, Group for Environmental Monitoring, Johannesburg.

Muller, Mike. 1999. “From Where Does the Money Come for Basic Levels of Service for Water Provision?”, paper presented at the Regional Workshop on Financing Community Water Supply and Sanitation White River, Mpumalanga, South Africa

November 26 - December 2.

Renzetti, Steven. 2001. “An Empirical Perspective on Water Pricing Reforms”, in Ariel Dinar (ed), The Political Economy of Water Pricing Reforms, Oxford University Press (for the World Bank), Washington, D.C.

RSA [Republic of South Africa]. 1994. Water Supply and Sanitation Policy White Paper, Cape Town: Government Printers.

—— 1995. Municipal Infrastructure Investment Framework, Ministry in the Office of the President and the Department of National Housing, 12 June. Mimeo.

—— 1997. White Paper on Water Policy. Pretoria: Government Printers.

—— 1998a. The Municipal Structures Act, No. 117. Pretoria: Government Printers.

—— 1998b. Department of Minerals and Energy, Draft White Paper on the Energy Policy of the Republic of South Africa, Pretoria: Government Printers.

—— 1998c. National Water Act, No. 36, Pretoria: Government Printers.

—— 2000a. Local Government: Municipal Systems Act, Pretoria: Government Printers.

—— 2000b. Draft: Local Government: Property Rates Bill. Pretoria: Government Printers.

—— 2000c. “Municipal Infrastructure Investment Framework (MIIF): Draft,” Department of Provincial and Local Government, February. Mimeo.

Whittington, Dale; Briscoe, John; Mu, Xinming, Estimating the willingness to pay for water services in developing countries: a case study of the use of contingent valuation surveys in southern Haiti, Economic Development and Cultural Change v 38 Jan 1990.

Whittington, Dale; Lauria, Donald T; Mu, Xinming, A study of water vending and willingness to pay for water in Onitsha, Nigeria, World Development v 19 Feb/Mar 1991.

World Bank. 1994. World Development Report: Infrastructure for Development. Oxford University Press, Washington, D.C.

—— 1999. Urban Property Rights Project Appraisal Document, Report No. 18245 PE, July 15, Poverty Reduction and Economic Management Unit, Latin America and the Caribbean Region. (accessed on May 8, 2002, at