Best practice guidance sets out what should be done but rarely explains how it will be paid for. Audit idealists demand 100 per cent compliance with funding rules whatever it takes. Inspectors report on shortcomings against best practice but never put a price tag on their recommendations. If there are costs involved, it’s up to college managers to find the money.

Models of good practice are proposed by national organisations regardless of cost, for example the recommendation in a Learning and Skills Development Agency report last year that every part-time student be interviewed before enrolment. All 4 million of them. Whether they need advice or not.

Similar proposals flow from the good people who design qualifications. Take the recent change to lecturer training qualifications. A proposal to raise standards by requiring the external observation of trainee lecturers significantly increases the costs of lecturer training courses. Who pays ? Someone else’s problem.

Assessment idealism is nothing new. Many NVQs failed because they made impossible demands on the time of assessors. Large organisations cope where they need the qualification. Others cut corners or took a simpler route.

One-third of NVQs never had any candidates.

The home of idealism in education is IT. For years, its proponents argued that e-learning would save money by replacing people with networks.

Students could access learning materials and wouldn’t need a lecturer. It hasn’t happened yet and there are manifest problems with quality. The new idea is blended learning - courses involving people and machines. Great if it works and if you have the money to design the programmes. Irrelevant for the short, customised courses needed to engage an employer or a small group of people.

The challenge to post-16 idealists is that time is money. If your good idea takes time for someone else to complete, you’re spending their money. If your idea is an instruction, you’ve given them no choice. And if your instruction reaches everyone in every college, you’ve spent millions.

Colleges might be able to handle the need to do more if they didn’t also have to pay more. We are in a period of rising pay costs. Lecturer pay costs could be 9 per cent more in September 2003 than one year earlier because of pay awards, pensions and national insurance.

There have been years of underfunding but rising costs present the Learning and Skills Council and colleges with a problem. The extra money coming in as income from the Treasury is consumed by pay inflation and the costs of best practice. It’s difficult to demonstrate added value when you’re running fast to catch up. This is not the only challenge for the LSC and colleges. An inheritance of malfunctioning systems complicates their relationship. The LSC is working hard to simplify and improve the way it funds colleges but this campaign will not bear fruit until 2004-5.

By then, it’s possible that the LSC will agree a plan with each college and base its funding on this plan. Funding will be agreed in advance and will not be taken away at the end of the year on the say-so of a teenage auditor.

Any shortfall by the college will result in adjustment to the plan and to the funding for the following year. This is a tried and tested approach to funding education, which can’t happen soon enough in the LSC world.

Unfortunately, it’s two years away.

In 2002-3, most colleges are still wrestling with a host of problems created by the current way of managing money. There’s a new set of funding rates but it’s difficult to interpret the reports coming out of the LSC software. There are so many rules, it’s difficult to know who has the answers.

There are targets and sub-targets but it’s unclear what happens if you hit one and miss the other. Responsive growth and the safety net may help but it’s unclear how either works. The funding claim auditors will be going away in future but they’re still here, now.

These are weak foundations for the reforms that the LSC wishes to introduce. Confusion in the present makes it harder to develop trust in the future. The new rewards for high performance make things more complicated in 2003-4 before they get simpler in 2004-5. And if all this was not enough, the skills strategy and the adult funding review promise radical change in the near future.

Compared with its predecessor, the LSC has achieved a remarkable amount of funding stability. Three-year funding agreements promise that this will continue to 2006 but you’d be unwise to bet on it.

Julian Gravatt is finance director of the City Lit, London