International workshop Venice - March 11-12 1999 Concepts and Paradigms of Urban Management in the Context of Developing Countries
	[ Workshop home page] [ Index of papers]

GILBERT Alan (University College London)

"The Future for Infrastructure Provision in Developing Countries"

Improving the quality of infrastructure and services continues to be one of the critical tasks for the new millennium. Despite having the technology, the managerial capacity and the resources to provide water, electricity, sewerage and buses for the world’s population, the twentieth century has seen the world fail to satisfy the basic needs of hundreds of millions of people. Far too many people lack basic services.

During the 1980s and 1990s public service agencies came under a great deal of criticism. They were not supplying services efficiently, they were not reducing prices and they were using monopolistic commercial practices. Many argued that a better answer lay through some combination of privatisation, the introduction of a more commercial logic and the delegation of more responsibility to local authorities. The shift occurred at a time when profound changes were occurring in the nature of macro-economic management. In most countries, the principles underlying government finance became less Keynesian, trade policy more liberal and the role of the state less interventionist. The ‘new’ public servicing strategy still constitutes ‘best practice’.

The purpose of this paper is to stimulate debate about the ability of the new approach to supply services to the poor of the Third World.

How serious is the infrastructure situation?

Infrastructure provision in the Third World faces innumerable problems. And, to judge from the many cataclysmic statements made in the press and even in academic journals, cannot possibly keep up with demand. Even in the middle-income countries, provision is inadequate. As the South African water agency describes it: "In a country with nuclear power, cellular telephones and vast inter-catchment water transfer schemes, more than 12 million people do not have access to an adequate supply of potable water; nearly 21 million lack basic sanitation" (DWAF, 1994: 1).

In most countries of Asia and Latin America, the evidence is much less conclusive. Indeed, in many places the figures suggest that infrastructure provision has more than kept up with demand. Water supply, sewerage provision and electricity generation have increased rapidly in most places. In South Africa, the situation seems to be improving but there are too many countries in the rest of Africa which have suffered badly. In Latin America, although the authorities in some cities have done a rotten job, most have improved service delivery despite rapid population expansion. Of course, there are innumerable problems with the way that demand is satisfied, the efficiency with which enterprises are run, the cost of the service, the pricing structures, and the environmental consequences of some of the methods employed. However, in most cities, infrastructure provision is not about to break down. Indeed, in recent years infrastructure provision has not only improved but improved impressively (Figure one). As the World Bank (1994: 1) point out: "During the past fifteen years, the share of households with access to clean water has increased by half, and power production and telephone lines per capita have doubled."

Figure One: GROWTH OF INFRASTRUCTURE PROVISION IN RECENT DECADES
(percentage decennial growth rate)

Sanitation and water data for 1975-90 period; power and road data for 1960-90 period.
Source: World Bank (1994: 5)

The main reason for this improvement is that huge amounts of money were invested both to improve the productive capacity of Third World countries and to improve social conditions. The World Bank (1994: 1) claims that "developing countries invest $200 billion a year in new infrastructure - 4 percent of their national output and a fifth of their total investment." Similarly, Crane (1994: 71) speaks of "the billions of dollars planned for urban water infrastructure investment over the next few years".

Large investments in servicing capacity were permitted by the fact that most Third World economies were expanding between 1950 and 1980. Virtually every economy in Latin America grew steadily and most Asian economies were expanding rapidly. Not surprisingly, given the close relationship between economic development and service provision, the quality and the coverage of service provision generally improved. Unfortunately, Third World economies do not always grow. Indeed, most countries of Sub-Saharan Africa have been in decline since around 1970, the lost decade of the 1980s hit Latin America badly, and recent events in East and Southeast Asia show that in the new, more competitive, global economy, continuous growth is the last thing any national or local economy can assume.

Economic decline has undoubtedly contributed to the disastrous servicing situation in much of Africa. As Nwaka (1996: 119-20) puts it: "The task of ensuring a healthy urban environment appears to have grown beyond the capacity and resources of municipal authorities, especially with the prolonged recession and the adjustment programmes which have weakened national economies and constrained the efforts to expand and maintain essential urban infrastructure and services." In Nigeria, "the economic recession and subsequent introduction of the structural adjustment programme has meant a declining allocation to urban infrastructure" (Ogu, 1997: 203).

Even where service delivery has apparently improved, we have to be a little suspicious of the very positive picture painted by the authorities. The fact that low-income settlements are linked to the mains does not mean that there is always water in the pipes! This has long been a problem in Venezuela and Peru but now seems to have afflicted Bogotá and Buenos Aires as well (Novaro and Perelman, 1994: 5).

The social consequences of limited infrastructure and services

The failure to expand infrastructure and servicing capacity sometimes impacts on the whole population. When there are cuts in the water or power supplies, the whole city is affected. However, it is clear that the poor are the main group to suffer from bad services. For a start the rich can cope better with cuts. They can buy private electricity generators and they have water pumps and large tanks to stock up when the supply is available. In addition, the rich receive services first. Private and public utilities alike are aware that while the rich can pay for services, the poor frequently can’t. The rich also have much more political clout than the poor. In consequence, it is a strange third World city where rich suburbs do not receive water or electricity (figure two).

Figure two: ACCESS TO INFRASTRUCTURE BY INCOME GROUP IN SELECTED COUNTRIES IN LATE 1980s

Source: World Bank (1994: 32)

By contrast, many poor neighbourhoods are not connected to the supply network, especially the new communities that spring up around the city (Glewwe and Hall, 1992; Gilbert, 1998). In addition, the service provided even after connection is often deficient. If the poor have limited access to services they also tend to bear the brunt of pollution. Water pollution is bad downstream from factories and higher income residential areas, the zones where many poor people are likely to live.

The consequences for health and welfare are obvious. Water-borne diseases are a serious problem in many third world cities. For example, those without easy access to piped water in the metropolitan areas of southern Brazil, "were 4.8 times as likely to die from diarrhoea than those with water piped to their house" (UNCHS, 1996: 135). It is not only disease that is the problem. As Crane and Daniere (1996: 204) point out: "...the difficulty in obtaining enough water is at least as prejudicial as water contamination to the health of the urban poor ... the high prices that the urban poor often have to pay for water diminish the amounts they have to spend on food and thus undermine their often precarious nutritional status."

The efficiency of public utilities in Third World countries

There is no question that public utilities in most parts of the Third World are less than efficient. Many are prone to corruption and tend to follow political objectives rather operating on the basis of either a commercial or a social rationale. Too many of these agencies make large financial losses, employ too many people, lose much of their electricity and water, and offer a generally unacceptable level of service (Roth, 1987).

At the same time, although it is rarely admitted, some public agencies in Third World countries are efficient. As Batley (1996: 748) puts it: "the better performance of private enterprise is by no means clearly established. Public refuse-collection systems (in Mexico and Zimbabwe) and education systems (in Malaysia and in Mexico) were among the best performers in the cross-national comparison. And purely public water supply systems were among the best-performing services overall." My own list of examples would include Brazil’s CVRD, Medellín’s service agency and Curitiba’s municipal authority.

According to the current conventional wisdom, however, the best way to improve service delivery is to change the way in which infrastructure and services are administered. For those on the ideological Right, privatisation and the removal of the damp hand of government, is the only solution (Stren, 1991; de Soto, 1985). The World Bank is often seen to be leading the campaign for privatisation in the Third World and is certainly eager to encourage the commercialisation of service delivery. The key task is to "manage infrastructure like a business, and not like a bureaucracy" (World Bank, 1994: 2). However, this does not necessarily mean it favours privatisation in every case. For as Mosley (1988: 138) argues "in its Structural Adjustment Loan policy conditions, the major instrument by which it has sought to reform policy towards public enterprises, it has concentrated very much more on the creation of competition and on efficiency audits for state monopolies than on divestment of those organisations."

In practice, the privatisation that has occurred in most Third World countries has often missed the major public utilities. With the exception of Chile, where water, electricity, telephones and transportation are now predominantly run by private companies, most Latin American utilities remain in public hands. Even where governments have tried to sell them, the private sector has often been reluctant to buy well-known basket cases. In South Africa, too, while there is a great deal of rhetoric about the need to privatise, little has so far happened. The vast railway network is still state owned and so are the main water, sanitation and electricity agencies. More signs of progress are apparent in the housing arena where rented municipal units have been given to the tenants but even there many problems have arisen (Emdon, 1993). As a result, the number of areas where private enterprise has taken over from the state is still rather limited.

The rhetoric for privatisation fails to recognise that many governments are sensitive about selling off key public utilities. Sometimes this is on grounds of principle, the state should run key sectors of the economy but often it is due to political opposition. In South Africa, for example, there was active union opposition to privatisation under both the Nationalist government and under the ANC. As Heymans (1991: 171) points out: "Vehement union protests in 1990 seem to have caused greater caution in government and private sector circles in the pursuit of privatization."

Sometimes, government reluctance is also due to the fear that the private sector will do a poor job. As Batley (1996: 749) points out "the introduction of private providers into urban service supply also brings risks: the possible unwillingness of contractors to deal with less-profitable areas and with poorer customers, pressure to raise prices in monopoly situations, inattention to positive and negative externalities, bankruptcy and the collapse of essential services, and difficulties of co-ordination between multiple producers. These risks represent costs which are transferred to government as the hidden cost of privatisation."

In the absence of resources, however, most governments do not need convincing that they should do less, they will do less anyway. This is the fundamental reason why so many governments have accepted the inevitability of a certain amount of privatisation. Decline has forced governments to cut back on expenditure which has prompted the realisation that government cannot now play the role that we perhaps should like it to play.

Nevertheless, in practice, the form of privatisation that has been most widespread in third world cities has not involved selling public utilities to the private sector. What is more typical is that the private sector has gradually filled the vacuum of service delivery in many Third World cities. "In crucial areas such as water supply, waste disposal, housing and mass transport, a genuine ‘privatisation’ of the supply of urban services and infrastructure has been occurring since the late 1960s" (Stren, 1988, 243). In the absence of an adequate response by the public sector, small-scale enterprise has been active filling the gap. "With the economic stagnation of the late seventies and early 1980s, this trend became more pronounced" (Stren, 1988: 218). A similar process has been occurring in the major cities of Peru (de Soto, 1989). Given the lack of investment in public transport, private taxis and minibuses begin to carry passengers. In the absence of water pipes, private water tankers move into the barriadas. In Buenos Aires, the past failures of the public company led to private initiative decorating office blocks with informal solutions telephone lines.

The application of commercial principles

Official policy these days is strongly influenced by World Bank thinking. In urban water and sanitation programs: "The Bank supports the recovery of enough revenues at least to cover operating costs and debt service. Cost recovery measures include connection fees, and block, flat-rate, or volumetric tariffs" (World Bank, 1992: 7). This approach also pervades policy in the regional development banks. For example the Asian Development Bank (1992: 23) argues that: "If subsidies are required because of equity considerations, they should be targeted and designed to be explicit, transparent and efficient." Third World governments often have little choice as to whether they wish to follow this logic or not because "the financial and economic reorganization of public utilities is one aspect of the overall question of structural adjustment in Africa. Such enterprises will have to achieve an operational balance that does not rely on the presumption that the State will bail them out financially or on tariff increases that would have an adverse effect on competitiveness and, as a result, on the growth prospects of the modern sector as a whole" (Lesueur and Plane, 1994: 383).

In South Africa, the Mandela administration has accepted this underlying logic. This is reflected in the first, slightly apologetic, post-election statement by the Water Ministry: "An insistence that disadvantaged people should pay for improved water services may seem harsh but the evidence indicates that the worst possible approach is to regard poor people as having no resources. This leads to people being treated as the objects rather than as the subjects of development; it generates proposals for unaffordable subsidies which tend to reach only those with influence, leaving the situation of the majority unchanged. Promises of free services for all have, in practice, usually resulted in some service for a few and little or none for most" (DWAF, 1994: 7).

Increasingly, this kind of argument is being strengthened by the assertion that people prefer to pay for services than not to receive them. As Stren (1988: 219) puts it: "Perhaps surprisingly, ... the urban poor have on balance benefited from this process of incremental privatisation. Not only are they more involved in the production and distribution of the services and facilities, but they gain access to urban benefits which (even for a price) they would not otherwise have enjoyed. Many urban residents (except for the very poorest) are not only willing, but even anxious to pay a market rate for what they perceive to be necessary services, delivered efficiently and regularly."

There is no doubt that that the financing of public utilities should be put on a firm footing. If it is not, capacity will be limited and the poor will be principal losers. However, there are clear dangers inherent in the application of user-cost pricing policies. First, it is difficult to charge consumers the full marginal cost of service provision when the cost is more than they can afford. Since many among the poor cannot pay for many of these services now, there is a clear limit to the amount that tariffs can be raised. If prices are increased then the danger is that services will not be used. In the field of transportation, fares remain unaffordable to many in certain African cities (Mabogunje, 1991) and, in Santiago, when deregulation allowed fares to rise dramatically, the proportion of journeys made on foot increased markedly (Figueroa, 1990). Extreme poverty clearly makes user-cost pricing less applicable in some circumstances and reduces the chance of efficiency savings. As Bell et al. (1993: 585) put it: because of high levels of unemployment "‘prices’ do not provide accurate signals about relative demand when consumers do not have sufficient income to exercise choice so the efficiency gains are not realised."

Second, there is the danger that existing consumers will react en masse against rising service charges. Certainly, many protests have occurred in Third World cities as a result of rising charges for infrastructure and services (Walton, 1989; Gilbert, 1998). This danger became manifest in South Africa during the 1980s. "When the first black local authorities were appointed in November and December 1983, they were under severe pressure to improve the level of infrastructure services. This could only be done by increasing rents, their single remaining source of income. As a result, there was an unprecedented rise in rents in townships throughout the country in 1984. These increases occurred during a period of falling real incomes, so that their impact was even greater. This led to a boycott of rents by residents, which began in the Vaal Triangle south of Johannesburg in 1984 and spread to more than 50 townships over the next five years" (Abbott, 1994). The payment situation seems to have improved little since 1994.

Third, there is the insidious danger that rising tariffs will increase corruption and illegality. In parts of Africa, Mabogunje (1991: 146) notes that: "there has emerged a counterculture in which urban households try to secure as many services as they can by illegal means." In Bogotá, where industrial electricity costs rose four times in real terms between 1977 and 1987, the rate of non-payment rose rapidly. A similar problem faced the city’s water company (Gilbert, 1990: 355).

Fourth, the scope for clawing back the cost of services is complicated when national currencies are devalued. In Bogotá, major loans were contracted during the 1970s to expand service capacity. With the rise in real interest rates in the late 1970s, and with the rate of devaluation against the dollar accelerating, servicing those debts became a major burden for the local agencies. The rising cost of debt servicing was passed onto customers. Of course it is reasonable that local clients should pay for the costs of expansion but insofar as those costs have risen dramatically because of external influences, should the lenders also not bear some part of the cost? Certainly, it is ironic if, as in the case of Bogotá, "huge debt repayments are threatening to undermine the only really effective public institutions" in that city (Gilbert, 1990: 359).

Fifth, although failure to apply user-cost pricing can certainly lead to unfairness, the application of commercial prices can have similar consequences. In South Africa, for example, the method of pricing for water had just this result. "The public funds available for water supply development have been invested mainly to assure that bulk supplies are available to those who can afford to exploit them. This has benefited farmers who can install pumps to take water from rivers to irrigate their fields, and municipalities which build plants to extract and purity water to sell to their citizens. Poor communities, for lack of both funds and organisation, have not been able to take advantage of their "right" to primary water supplies" (DWAF, 1994: 4).

Ultimately, the justification and feasibility of user-cost pricing depends on the wider environment. If the inequality of income, the political situation or the poverty of the population is sufficiently acute, then user-cost pricing will simply not work. In South Africa, it did not work in the Black Local Authority areas under apartheid. The social inequalities and the costs levied on BLAs were such that they removed any possibility of balancing accounts (Bell et al., 1993: 583). The boycotts against paying rent and service charges are continuing even under the Mandela government. The Masakhane campaign for everyone to pay, has so far met with little success (Lodge, 1996). Botes et al. (1996: 463) comment on experience in Bloemfontein where: "A water connection deposit of R50 per stand was required by the local authority. Despite promises to the contrary, very few people paid their deposits and in the end the Trust had to do so. Water was supposed to be paid for according to usage, but getting people to meet their commitments became a major problem."

Will greater efficiency through private provision improve infrastructure provision?

Market efficiency does not necessarily equate with social efficiency. Why should commercial suppliers provide infrastructure to the poor? If suppliers concentrate on the commercial attractiveness of particular segments of the market, they will seek to provide increasingly sophisticated services to the better off and offer nothing to the poor. Providing small quantities of water to poor households across the city is not appealing to a commercially oriented supplier. In most developed countries, at least most poor households can afford to pay the charges for water, electricity, sewerage and telephones. In many less-developed countries, however, the situation is different; many, perhaps most, consumers cannot afford to pay much for services. For the supplier, therefore, it is not efficient to supply them.

Supplying poor households only becomes ‘efficient’ when account is taken of the reduced morbidity which accrues from providing a safe water supply. Even then, the benefits may be undermined if some kind of drainage and sewerage system is not provided simultaneously. In short, the market has always had problems with external economies and diseconomies.

One counter to that argument is that in practice the poor can afford services. The arguments of observers like De Soto (1987) and Cohen (1990) suggest that the key problem is not the lack of effective demand but overregulation; the poor are willing and able to buy services provided that suppliers are allowed to supply them. Hence Cohen (1990: 54) argues: "If many urban households were already receiving water through private water sellers and not through individual water connections to the public distribution system, why not further legitimize private water companies on a larger scale? Experiments in Abidjan and Santiago de Chile have successfully demonstrated the feasibility of this option." Personally, I would be more convinced if the vacuum were being filled by charities or non-governmental organisations of a co-operative kind. This would seem to be a less problematic development and one in keeping with the praise being heaped on such organisations in so much of the literature.

Regulation

In the past, the state reacted to market failure by setting up public companies to provide services. Today, the conventional wisdom is that the state should intervene only by regulating the suppliers. If governments insist that every part of the city be provided with services, then commercial providers will be forced to do so. Unfortunately, in many a Third World city, it is not quite that simple.

First, governments are not that efficient as regulators. Even in Britain, the record of regulators since privatisation has not been unblemished. If that is the case in Britain, how much more problematic is the issue of regulation in Bogotá or Cape Town? For, if Third World governments are quite good at imitating the laws of First World societies, they are much less good at applying and monitoring those laws. If you doubt me look at any of the wonderfully liberal constitutions drawn up when Latin American countries became independent in the 1820s and compare the fine words with everyday reality. Many Third World societies are notorious for the amount of corruption and intrigue that goes on. Setting up a regulatory mechanism may merely create a new arena for bribery.

Second, even if regulators insist that private operators comply with stiff performance indicators and threaten to remove licences if the service is deficient, will that improve delivery to the poor? The answer is only sometimes. When the profit margin is too low, stringent regulation may simply persuade the supplier to cease operations. A clear example of the problems that result comes from Benin City, Nigeria, where, "with rising inflation, and pegged service charges, private operators began to find it difficult to maintain their vehicles and subsequently began to withdraw" (Ogu, 1997: 212). We are back to the issue of effective demand. Without a subsidy, the supply will not be forthcoming. But, under the current mode of thought and present economic circumstances, there is little chance of a subsidy.

Third, there is a strong ethos at the moment to cut back on the nanny state. As I observed above, the argument in much of the literature is that the state already intervenes too much. According to de Soto and Cohen, private enterprise should be encouraged to operate in low-income communities. Reform and even removal of the regulations would automatically formalise the informal. Modification of the regulations would allow small-scale entrepreneurs to perform functions that are currently the preserve of other institutions. I have no problem with the argument that the private sector be permitted to provide services providing that the state regulates its performance. Service delivery by private companies in a socially aware and environmentally aware context is one thing; in the current days of free markets and the rolling back of the state, the same strategy may be less socially productive.

Participation of the community to assist delivery

Great play is currently being made of the goal of taking government closer to the people. Decentralisation of government is de rigueur in many countries. In Chile, municipalities were given much greater power under the Pinochet government. In Colombia, the new constitution of 1991 warmly embraced both decentralisation and participation. And, in South Africa there have been strong moves towards decentralised power. Such a move is strongly encouraged by the Washington consensus and forms part of the planned demolition of ‘big’ government. To what extent decentralisation has encouraged greater efficiency and to what extent it has led to people being consulted more about their needs is debatable. The danger with decentralisation, delegation and consultation in Latin America and South Africa is that it may turn into political populism. By this I mean that generous promises to the people cannot be implemented; they are neither replicable nor sustainable.

However, greater public participation also provides a different kind of potential solution. Services can be supplied more cheaply if the poor are involved either in the provision of services or in specifying the form of the service to be provided. The poor should be encouraged to participate in the design, provision and maintenance of infrastructure and services. Such an approach has been strongly advocated by the multilateral agencies. For example, the World Bank (1992: 7) argues that: "Difficulties with cost recovery in slum-upgrading projects can be reduced by promoting extensive community participation at an early stage, so that there is broad agreement from the outset about goals and methods." Similarly, Lee (1995: 1) argues that: "In many cities, it is not so much a lack of demand for basic infrastructure and services that is the problem, but rather a lack of institutional capacity at the community level to support the provision of affordable and effective services, and a reluctance of professionals and bureaucrats to develop and accept alternative local innovations."

There is certainly no shortage of evidence that the poor can and do participate in service delivery and in neighbourhood improvements. Indeed, on reading some of the current literature the poor are so active in community projects or in protesting on the streets that one wonders how they have any time left to earn their living (Annis, 1987; Seddon and Walton, 1993)! Unfortunately, my own and others’ experience suggests that the amount of community participation is much more limited (Gilbert, 1988; Mainwaring, 1989; Munck, 1993). Communities are riven by social and political divisions and many initiatives are undermined by mutual distrust. Settlements which contain large numbers of tenants are much less likely to participate in community improvements; the tenants are reluctant to improve services in case their rents are raised and, in any case, have their eye on a plot of land where they can build their own home and participate in community schemes that will benefit their own family. Other settlements suffer from the fact that the community does not perceive the neighbourhood as home. A common African dilemma is described by Ogu (1997: 210): "many of the migrants are not committed to urban life-styles. Since these non-indigenous urban residents usually see their stay in the study area as temporary, they will not be keen to involve themselves ...."

In places, even where community involvement might once have been common, it has been undermined over the years. This is very common in countries such as Venezuela and Nigeria where government patronage funded by oil royalties has produced an attitude where: "Many residents in the city perceive urban services as a government responsibility, which they are entitled to enjoy free of charge. This belief is often encouraged by local council politicians who, in order to win votes, try to create the impression that the services are intended to be free" (Ogu, 1997: 210).

Brown and green agendas

Nwaka (1996: 120) points out "there is growing support in the literature for a distinction between the "green agenda" for long term environmental security, and the "brown agenda" of environmental issues associated with the immediate problems of survival and development in the towns and cities of the developing world." There are even signs of this on the ground, for example, in the Settlement Infrastructure Environment Programme (SIEP) of UNCHS which arose out of Agenda 21. Unfortunately, in Third World cities there is potentially an inherent conflict between satisfying the goals of widening service coverage, greater cost-effectiveness and preservation of the physical environment.

An example of the kind of conflicts that may arise are indicated by Novaro and Perelman (1994: 7) in Buenos Aires: "the strategy followed by the water and sanitation agencies gave priority to extending the network, even when this undermined the quality of supply and polluted the environment." Such an outcome is always likely when commercial agencies are trying to increase the number of subscribers. It is also likely when politicians are under pressure from electorates to deliver on promises of taps, drains and electricity lines.

Can technological advance save us?

The industrial revolution brought with it the belief that technology can improve the human condition. That belief was partially vindicated during the twentieth century, most of us now live much longer lives. What improved technology has not so far delivered is a consistent reduction in poverty.

The critical problem with technology is that it is all too often put to the wrong use. Frequently, technological progress occurs fastest in areas which are the most profitable. Sometimes, profitability means the same as socially productive, for example, the production of cotton and later synthetic clothing and of certain medicines generally brought much more good than evil. Too often, however, technological progress has catered principally for higher income groups without bringing widespread gains for the majority of the human race. Arguably, supersonic aeroplanes and most ‘advances’ in pharmaceutical and medical practice fall into this category. Whether the poor benefit from improved technology therefore depends on the nature of the innovation and the economic and social context in which it is produced and sold.

Nowhere are the contradictions between technological advance and social progress better demonstrated than in the arena of transport provision. Technology has brought major improvements in our ability to move about, indeed, humans have never been so mobile. We can travel supersonically, we can travel in comfortable and reliable cars with stereo music systems, we can travel cheaply across the Atlantic and to Australia. Despite these technological advances, the transport situation in most cities has probably deteriorated. The key problem is that rates of car ownership have been growing too fast. Road capacity has increased but much more slowly than the number of cars; in part because of budget constraints, in part because of the sheer increase in the numbers of vehicles, and in part because it is simply not possible to create more and more roads in most urban areas. The consequence of this trend is obvious: vehicle use increases, traffic congestion worsens, the quality of the air deteriorates.

Rather than dealing politically with the basic problem, the private car, technological advance has been used to develop other ‘solutions’. In Caracas, Medellín, Mexico City, Rio de Janeiro and Santiago, the authorities have taken the capital-intensive option of constructing new ‘metro’ transport systems (Figueroa, 1996). While ‘metros’ are very popular locally and help more people to get to work, they do not reduce traffic congestion for long. Meanwhile, as in Britain, other forms of public transport are allowed to deteriorate and improvements to bus systems are few and far between. Rhetorical statements are made about the desirability of cycling and walking, but insofar as priority is given to anything it is given to building metros and motorways and to encouraging greater car ownership. So long as governments are reluctant to say ‘no’ to the private car lobby, conditions will deteriorate. My fear is that the combination of growing private car use, economic expansion, the ethos of liberalisation, distrust of the state and the failure to improve the bus system will lead to an economically expensive solution; the building of a ‘metro’.

If technological advance in the transport arena promises us little, perhaps, improvement in the field of telecommunications promise more. Certainly, cheaper telephones are likely to bring benefits to many although it is less than certain what proportion of Africans will be using mobile phones by the year 2050. We can speculate on how other advances in technology will benefit the poor in most Third World countries. My fear is that the huge potential improvements that technology could bring will, as usual, be substantially wasted. It is not that there will be no progress, only that there will be far too little.

CONCLUSION

It would be convenient to claim that the new conventional wisdom is all wrong and that we need a revolution in thinking that will sweep it away. I do not believe that. Previous conventional wisdoms all had many vices and much of what the current ethos offers is certainly sensible. However, we must be very wary of expecting very much. The key problem is that the new conventional wisdom is not actually new. In fact, many of its theoretical premises come from the nineteenth century. Insofar as modern economics preaches that the modern economic system should be based on free trade, on foreign investment, on limited state intervention and on market forces, it does not seem very different from the writing of nineteenth century free traders.

The difficulty for service provision is that improvements have to occur within a globalised and less regulated economic world. While it is true that we are more aware than ever of the dangers that the world faces, global warming, poverty, disease, etc., we seem always to have our hands tied behind our backs when it comes to solving these problems. Technology should improve our ability to solve the problems but does not often do so. To use a controversial example, humankind uses technology to produce nerve gases, and cruise missiles to attack the factories producing those gases, rather than using technology to help reduce poverty in Iraq. Technology is all too often used for precisely the things it should not be used for.

If service agencies are now managed on the basis of catering for demand rather than merely increasing supply, this may well mean that utilities will demonstrate "renewed sensitivity to changing user needs via more accurate demand profiles and more flexible network management through refined customer and service differentiation" (GEC briefing) The danger, in a Third World context, is that privatised utilities might be tempted to provide the rich with services that are profitable and not provide the poor with services that the latter need but cannot afford. Regulation will not provide the answer because Third World regulators will be even less effective than regulators in Britain. Community participation will not fill the void because under the new conditions of flexible accumulation, everyone in the shanty towns will be too busy working to pay for the higher prices of unsubsidised services. Insofar as they get together as communities to respond to the challenge of service provision, it will be to protest about the rising cost of servicing, the inefficiency of the providers, and the failure of government promises. What I am suggesting is that privatisation and the unleashing of market forces may produce a form of service delivery that is likely to damage the quality of life of the poor and the lack of regulatory control is like to further damage the environment. Companies will expand production, higher incomes will increase private consumption, there will be too little implementation of the green agenda and the poverty of the shanty towns will impede progress in satisfying the brown agenda. In the end, the combined forces of globalisation, free enterprise, weaker government controls and public anarchy will lead to a situation where almost no-one is happy.

If the new conventional wisdom about service delivery were motivated primarily by a wish to help the poor, to involve the poor in identifying priorities, in reducing environmental damage and generally to improve the quality of life, then the shift would undoubtedly be a ‘good thing’. But, remember the recipe that has been emerging from the World Bank (1994: 8): "To reform the provision of infrastructure services, this Report advocates three measures: the wider application of commercial principles to service providers, the broader use of competition, and the increased involvement of users where commercial and competitive behavior is constrained." I am unconvinced that this will "produce three types of gains: reduction in subsidies, technical gains to suppliers, and gains to users." My main reason for doubting the outcome is that the motivation behind the shift in policy is not primarily to help the poor. Principally, I believe it has occurred because governments are looking for ways of cutting government expenditure.

As a result, the declared aims of service delivery cannot be satisfied by the methods being employed currently to provide those services. Thus privatisation is incompatible with improved environmental sustainability, cost effectiveness too often incompatible with poverty reduction, reductions in taxation alien to the expansion of services. The rhetoric that has developed about social safety nets, decentralisation and sustainability is designed to cover up the inability of either the modern or the post-modern state to satisfy most of the demands of society for services. This is because societies now espouse a series of mutually incompatible desires. They want more and better services, they want less pollution, they want to consume more but critically they want to pay less taxes and less for key public services. To paper over these mutually incompatible desires, it has been necessary to create a new mythology: the state will create the conditions for improved service delivery by doing less and in the process will help the environment, reduce poverty and cut taxes.

In 1991, I made the following pessimistic prediction: "by about 1995, government regulations will have been cut back a great deal and people will have been left even more to their own devices. No doubt this will have brought some benefits, especially for the affluent minority who can afford to pay for quality services. But the majority will have discovered how exploitative most illegal subdividers are, in what a dangerous state most private companies run their buses, and how few private entrepreneurs are prepared to provide services to the unprofitable poor. Most will also have realised that government has maintained the urban fabric even less effectively than previously. At this stage there will be once again a great clamour for public intervention, regulation and control" (Gilbert, 1992). I was obviously foolish in putting a precise date in the text, but am I wrong in still believing that my basic fear will sometime be proved correct?

READING

Abbott, J. (1994) ‘The decline of the urban management system in South Africa’, Environment and Urbanization 6, 201-13.

ADB (Asian Development Bank) (1992) Annual Report 1992. Manila: Asian Development Bank.

Annis, S. (1987) ‘Reorganization at the grassroots: its origins and meaning’, Grassroots Development 11, 21-25.

Batley, R. (1996) ‘Public-private relationships and performance in service provision’, Urban Studies 33, 723-51.

Bell, M.E., Dearborn, P.M. and Hunter, R. (1993) ‘Financing the post-apartheid city in South Africa’, Urban Studies 30, 581-92.

Botes, L.J.S., Stewart, T. and Wessels, J. (1996) ‘Conflict in development: lessons from the housing initiatives in Freedom Square and Namibia Square’, Development Southern Africa 13, 453-68.

Cohen, M.A. (1990) ‘Macroeconomic adjustment and the city’, Cities, 7, 49-59.

Cook, P. Kirkpatrick, C. (1988) ‘Privatisation in less developed countries: an overview’, in Cook and Kirkpatrick (eds.), 3-44.

Cook, P. and Kirkpatrick, C. (eds.) Privatisation in less developed countries, Harvester Wheatsheaf

Crane, R. (1994) ‘Water markets, market reform and the urban poor: results from Jakarta, Indonesia’, World Development 22, 71-83.

Crane, R. and Daniere, A. (1996) ‘Measuring access to basic services in global services: descriptive and behavioral approaches’, American Planning Association Journal 62, 203-21.

Damián, A. (1992) ‘Ciudad de México: los servicios urbanos en los noventas’, Vivienda 3, 29-40.

Davis, M. (1996) ‘South Africa’s electrication programme: progress to date and key issues’, Development Southern Africa 13, 469-84.

de Soto, H. (1989) The Other Path, I.B. Taurus. (First published in Spanish in 1986).

Emdon, E. (1993) ‘Privatisation of state housing: with special focus on the greater Soweto area’, Urban Forum 4, 1-13.

ESRC (Economic and Social Research Council) (1996) Demand side management and urban infrastructure provision, Briefings from the GEC Programme 6.

Figueroa, O. (1990) ‘La desregulación del transporte colectivo en Santiago: balances de diez años’, Revista EURE 49, 23-32.

Figueroa, O. (1996) ‘A hundred million journeys a day: the management of transport in Latin America’s mega-cities’, in Gilbert, A.G. (ed.) The mega-city in Latin America, UNU Press, 110-32.

Gilbert, A.G. (1987) ‘Forms and effectiveness of community participation in squatter settlements’, Regional Development Dialogue 8, 56-87.

Gilbert, A.G. (1990) ‘The provision of public services and the debt crisis in Latin America: the case of Bogotá’, Economic Geography 66, 349-61.

Gilbert, A.G. (1992) ‘Third World cities: Housing, infrastructure and servicing’, Urban Studies, 29, 435-60.

Gilbert, A.G. (1998) The Latin American city, Latin America Bureau, revised edition.

Glewwe, P. and Hall, G. (1992) Poverty and inequality during unorthodox adjustment: the case of Peru, 1985-90, Living standards measurement study Working Paper No. 86, The World Bank.

Heymans, C. (1991) ‘Privatization and municipal reform’, in Swilling, M., Humphries, R. and Shubane, K. (eds.) (1991) Apartheid city in transition, Oxford U.P., 151-73.

Lee, T. (1994) ‘Water management in the metropolitan areas of Latin America’, in UNCRD, 29-40.

Lesueur, J-Y. and Plane, P. (1994) ‘Human resource management and the restructuring of public utilities: water and electricity in Africa’, International Labour Review 133, 372-385.

Lodge, T. (1996) Local government politics, Optima, February, 37-39.

Mabogunje, A.L. (1990) ‘Urban planning and the post-colonial state in Africa: a research overview’, African Studies Review 33, 121-203.

Mainwaring, S. (1989) ‘Grassroots popular movements and the struggle for democracy: Nova Iguaçu’, in Stepan, A. (ed.) Democratizing Brazil: problems of transition and consolidation, Oxford University Press, New York, 168-204.

Mosley, P. (1988) ‘Privatisation, policy-based lending and World Bank behaviour’, in Cook, P. and Kirkpatrick, C. (eds.) Privatisation in developing countries: a survey of the issues, New York and London.

Munck, R. (1993) ‘After the transition: democratic disenchantment in Latin America’, European Review of Latin American and Caribbean Studies 55, 7-20.

Novaro, M. and Perelman, P. (1994) ‘La provisión de agua en el Gran Buenos Aires’, Boletín Medio Ambiente y Urbanización 49, 3-15.

Nwaka, G.I. (1996) ‘Planning sustainable cities in Africa’, Canadian Journal of Urban Research 5, 119-36.

Ogu, V.I. (1996/7) ‘Enabling strategies and sustainable infrastructure development: focus on Benin City, Nigeria’, Regional Development Strategies 3, 199-217.

Riedel, J. (1993) Public investment and growth in Latin America, Inter-American Development Bank Working Paper Series 134.

RSA (Republic of South Africa, Department of Water Affairs and Forestry) (1994) Water supply and sanitation policy: White Paper. Water - an indivisible national asset, Cape Town.

Roth, G. (1987) The private provision of public services in developing countries, Oxford University Press.

Seddon, D. and Walton, J. (1993) Free markets and food riots: the politics of global adjustment, Blackwell.

Stren, R. (1988) ‘Urban services in Africa: public management or privatisation?’, in Cook and Kirkpatrick (eds.), 217-47.

UNCHS (United Nations Centre for Human Settlements) (1996) An urbanising world: global report on human settlements 1996, Oxford University Press.

UNCRD (United Nations Centre for Regional Development (1994) Enhancing the management of metropolitan living environments in Latin America, UNCRD Research Report Series No. 1.

Walton, J. (1989) ‘Debt, protest, and the state in Latin America’, in Eckstein (ed.) (1989) Power and popular protest: Latin American social movements, University of California Press, 299-328.

World Bank (1992) Poverty Reduction Handbook, Washington D.C.

World Bank (1994) World Development Report 1994: infrastructure for development, Washington D.C.