The International Policy Centre for Inclusive Growth (a UNDP-funded thinktank, based in Brazil) has published an excellent 35 page ‘Poverty in Focus’ on public v private provision of basic utilities, especially water and sanitation, (but also touching on electricity). Some highlights from the overview, based on a series of country case studies:
‘Rapid urbanisation and informal settlements pose particular problems for water provision. The number of residential water connections has fallen in 
most unplanned urban settlements in the past decade. Large-scale private providers cannot resolve a number of obstacles without imposing exorbitant tariffs to cover costs. There are serious doubts about the potential gains of both privatised network utilities (where planning and development challenges persist) and small-scale service providers (because of pricing and quality issues). Ultimately, these concerns can be resolved by investing in the expansion of the public water and sanitation network.
The debate on private versus public provision of utilities is complex, but the guiding principle for the kind of provision preferred must be the initial level of access to water, sanitation and electricity. Where access is already high in developed and middle-income countries, privatisation may yield productive and dynamic efficiencies. Private providers have incentives to improve overall performance through new techniques and novel management processes.
Where access to utilities is low and the focus is on increasing coverage of the poor in low-income countries and neighbourhoods, public provision makes sense. This is because of problems associated with affordability, how much cost recovery can be pushed, and regulatory capacity. The persistent challenge, however, is financing investment outlays. The options are reducing system losses such as water leakages; improved billing; domestic resource mobilisation; and external financing (both donor and private bond/equity financing).
Historical experiences are particularly enlightening. Privatisation had been relatively successful in the United Kingdom [I can think of a few Brits who might take issue with that, especially on the railways, but never mind] and the United States, because these countries embarked on private utility provision after achieving 100 per cent access to water and electricity by the 1980s. The sewerage systems in Europe, the United States and Japan were not developed through full cost recovery from users; they were paid for by distributing the costs among the public, using taxation and cross-subsidy.
And the case studies?
In Bolivia, the private concessionaire and the government agreed on 
coverage targets to provide universal access in the city of La Paz and 82 per cent coverage in El Alto by 2001. The poor’s access to water connections increased, but the private company could not meet the targets. Inevitably, the limits of cost recovery and profitability had been reached. The tariff increases needed to connect the additional poor consumers were so high that they sparked public outrage (see pic).
Similarly, private concessionaires in Argentina entered into a contract with the government to increase water metering up to 100 per cent. Fees were imposed for the installation of the meters and tariffs were increased. The result was intense public protest. In Mali and Senegal the poor have not benefited from privatisation, simply because they were not connected to the grid in the first place. The tariff hikes after privatisation affected them indirectly as a result of economy-wide effects. The concessions in Argentina, Bolivia, Mali and Senegal have all been terminated.
Contract cancellations and renationalisation are often the result of a policy that transfers risk to governments and end users. In industrialised economies, the transfer of risk to the private sector is considered essential if efficiency gains from privatisation of the delivery of basic services are to reach end users. In Sub-Saharan Africa, however, the emphasis is on reducing the risks faced by the private sector in order to encourage private investment. The upshot is always exorbitant tariffs and neglected infrastructure.
One reason why private participation in the water sector has been successful in Brazil seems to be the transfer of investment risk. Contracts with the various government entities at the state and municipal levels clearly outlined the investment obligations of the private operators, particularly in low-income areas. The private operators had invested about U$500 million by 2004. The positive outcomes in Brazil are related to contract design… Most contracts stressed investment obligations, something relatively easy to monitor.