Alison Wolf: Ministers should stop treating adults as stupid children

British governments are convinced, in the face of overwhelming contradictory evidence, that they can predict the future. You might think they would be disabused by the financial crash, our limping economy, and the yawning gulf between the Treasury's expectations and its confident predictions in times of plenty.

Not a bit of it. They still think they know what will happen, for years ahead, far better than the rest of us. This conviction drives the whole of post-compulsory education policy and funding other than for the universities. It explains why governments treat adults and employers as stupid children, telling them exactly what to learn, and how to train.

This needs to change. Money must follow learners, not government contracts, and so create a genuinely demand-led system. Governmental delusions have laid further and adult education waste. They have also created an appetite for change. This is true even though professionals and institutions are exhausted by decades of interference, bureaucracy, and the endless restructuring of quangos, inspectorates and government departments.

How do I know? Because of the response to my report How to shift power to learners, written for the Learning and Skills Network. I propose new financial and regulatory structures that allow colleges to choose exactly what to offer, including their own awards; and do away with most existing quangos while still safeguarding quality. We must also give learners direct access to education subsidies through, critically, the creation of genuine learning accounts.

The government's latest wheeze is a new form of central planning relabelled as "skills activism". So, I am not expecting ministers to endorse this tomorrow. But it is clear that support for structural reform exists among educators, awarding bodies and the quangocracy itself.

Learning accounts – real learning accounts – would operate just like any other bank, building society or post office account, but be used only to pay for learning. That could be anything from bricklaying to maths A-level, programming to philosophy: the learner chooses. People would pay money into them, and the government likewise.

The obvious analogy is Gift Aid, where the government tops up our gifts to charity with additional funds. Learning top-ups would go into individuals' own accounts, at levels that varied with individuals' circumstances, and with the size of the further education budget. It would be for the account holder to decide when to use the money – again, like any bank account. And with the learner in control, a far simpler system of inspection and accreditation of learning providers could also be achieved.

There is a myth in Whitehall that learning accounts cannot be used because they are intrinsically susceptible to fraud, and because spending cannot be controlled. Neither is true, and both can be traced to the unhappy story of the first New Labour administration's Individual Learning Accounts. Launched in 2000, these were closed down "because of fraud" in 2001; and very large numbers were opened in their short lifetime. In fact, as the education select committee established, there was not actually much fraud at all. There was, rather, a very badly designed scheme, launched too fast, and with almost non-existent quality control.

The Commons committee was very keen to see ILAs return. What we have had, instead, is a series of centrally planned initiatives that makes it hard to believe that Labour once trusted individuals to make their own decisions. Since 2001, the government has channelled more and more of the FE budget into the Train to Gain programme that delivers qualifications in the workplace. It largely involves formal accreditation of existing skills rather than training anybody, and was recently panned by the National Audit Office.

More generally, further education and training increasingly revolve around detailed contracts, issued by government to providers. Trainers agree to deliver qualifications, designed to government specifications, at the levels, in the quantities, and in the sectors the Government thinks we need.

In other words, government knows what the economy of today and the future need. Only they don't. Many of these qualifications bring no financial benefit to the people who obtain them; and the Government's own surveys find an oversupply of qualifications in most of the labour market.

Central planning never works – this time is no different. Let individuals decide what learning might serve them best, now and for the future. They will do a far better job.

The writer is professor of public sector management at King's College London.