The standard arguments for privatisation are that competition produces greater efficiency and lower prices, and that governments should not carry the risk associated with running these enterprises.
Well, lets examine the claims. There's a good article on deregulation in Australia by Tim Anderson from the University of Sydney. Considering the claims of price savings in telecommunications since deregulation, he concludes:
...the evidence to suggest that actual competition has generated market efficiency, or greater benefits for consumers, is very weak. However there has been a gain for corporate efficiency, as large private corporations have been allowed a new share of a highly profitable market.
And in a particularly ludicrous twist, for the sake of efficiency, the rolling out of fibre-optic infrastructure in 1996, demanded the rolling out of two identical sets of infrastructure:
It does seem extraordinary that, on the one hand, the new technology of fibre optics delivers the potential efficiency of combining several functions (telephony, data connections and cable television) in the one network, yet on the other hand an economic policy demands the inefficiency of immediately duplicating that physical network. The head of the Australian Competition and Consumer Commission, Professor Alan Fels, defended the dual cable layout, claiming that competition would push prices down. However telecommunications expert Dr Peter Troughton condemned the duplication as "crazy" (Walker 1996: 4). Authorisation of the dual cable layout is explicable, though, as state support for the strategic move by Optus to secure its share of the market, by ownership of vital infrastructure.
On the question of risk, where's the risk? Are elecrticity and water likely to become suddenly unfashionable? As John Ralston Saul has pointed out, private corporations want public assets precisely because there is no risk associated with them. These are guaranteed cash cows, even more profitable if governments can be pressured into dropping the requirements that these utilities provide a minimum level of service to the whole population.
The main issue is who is responsible to whom? A private corporation answers only to it's shareholders, so it's definition of efficiency is likely to be very different from it's customers and the general community. In fact, corporate efficiency is the antithesis of market efficiency (remember, economics was born in an age before corporations).
Efficient markets are supposed to reduce the margin between the cost of producing a good and it's price in the marketplace. An efficient corporation however, seeks to maximise it's margins to produce the greatest possible return on investment for it's shareholders. So it keeps prices as high as the market will bear, while shifting costs elsewhere. It transfers costs to it's employees, by shedding staff and increasing working hours (that's called raising labour productivity), or to it's customers and the general community, by lobbying to remove minimum level of service requirements, environmental requirements, and so on. This is called deregulation, although of course it's just re-regulation, shifting control over what goes on from the government, which may be somewhat democratic, to corporations, which are as Noam Chomsky rightly says, as totalitarian as any institutions ever devised.
The public, meanwhile wants affordable goods, reliable services, and by the way, reasonable hours of work and good pay. If you measure performance by these criteria, "good corporate governance" is staggeringly inefficient in just about any enterprise you care to think about, certainly in the provision of essential services which are, usually, natural monopolies.
The ultimate answer is that all forms of industry should be democratically run. This means that the institutions concerned should answer firstly to the people who work there, secondly to their customers, and thirdly to the communities that are affected by their operation. There is no place for shareholders, because shareholders are market-distorting, as seen above.
How this is to work in practise is a tricky matter, and I wouldn't presume to guess at how people will want to do things. It's not a job for manifesto writers or "vanguard parties". The immediate task is to put into place the forms of organisation that will allow democracy to function in the industrial sphere. Here's one of them.