In the view of the City, the growth prospects for companies in public-sector outsourcing are good. Support services have replaced technology, media and telecoms as a top sector for investment. Construction and computer companies have reinvented themselves as providers of managed services.

Despite its reputation, the college sector is not particularly fertile ground for the private sector. The extent of private-sector involvement in publicly-funded further education has not increased much in 10 years. Private and charitable training providers have the lead role in work-based learning, but core public funding goes to public-sector colleges. Colleges use private-sector property and contract caterers but they have not really ceded control of their core business in the classroom. In the 1990s, many colleges used agencies to supply part-time teachers and franchised provision to companies.

Public controversy over these developments forced political, funding and legal changes. These, in turn, made franchising and staffing agencies less attractive. The trend looks set to continue. Ministers talk about quality not cost. Money is allocated to improve the pay and conditions of directly-employed teachers.

There are other less visible obstacles to college outsourcing. VAT adds 17.5 per cent to the costs of any service but colleges cannot add VAT to their fees or public funding. Post-16 education is unique in the public sector in facing a privatisation tax. Hospitals and local government can both reclaim VAT on outsourced services. Large companies also find they cannot spread their costs over large contracts because colleges are relatively small. The typical budget of the average college - pound;10-15 million turnover, 1 per cent operating surplus - leaves little room for large companies to make acceptable profits.

Thus private involvement in post-16 has been a bit of a cottage industry. Hundreds of people made redundant by their colleges have reinvented themselves as self-employed consultants. Hundreds more have made the same change voluntarily to increase their pay or independence. A few of the cottages have proved to be goldmines for their owners. The EMIS computer company was sold to Capita in 1999 for pound;20m. Network Training was sold to the Tribal Group for pound;11m. Another buyer is the Protocol group, which owns ELS. Within six months in 2001, it bought several training providers including Spring Skills, Tektra and the Initial staffing agency. Between them, Capita, Tribal and Protocol have bought up most of the FE cottage industry in less than three years. Perhaps the next stage is full-scale property development.

This buying of companies is a sign that some people believe there is money to be made in post-16 education. However, it is also a sign that others would rather sell up and exit stage left. Anyone working in the sector has to contend with low margins and political risk. Although the Government complains about the inflexibility of public-sector providers, it can rely on colleges and charities to look beyond the bottom line when making decisions on hard-to-reach students or useful-for-the-economy courses. There are no known cases of companies offering to buy public-sector colleges and run the business instead on the same terms.

A dramatic shift towards the private sector requires a curriculum innovation that pays. If a private company could use technology to change the basic economics of education, it could make money where colleges currently spend it. The optimists put their faith in e-learning but the products on the market so far are inaccessible electronic books or expensive courses with vast development costs. Most dot-com training businesses disappeared. The government-funded Learndirect survives but only by offering a limited curriculum to large numbers.

Until we get a curriculum innovation that pays, the college market will stop at the classroom door.

Julian Gravatt is director of finance at the City Lit, London.