Widening participation, the government's most cherished ÖйúAƬ policy, has a fine ring. Who could oppose improving opportunities for students from families with low incomes? As this week's Organisation for Economic Co-operation and Development survey shows (page 14), getting a degree is the best way for people to improve their life chances.

But making it happen is something else. The easy way to widen participation is to increase supply. There are no losers. Women were the main gainers from expansion in the 1960s, ethnic minorities have seized the opportunities opened up in the 1980s and early 1990s. But expansion is expensive - or if done on the cheap, shoddy. Expansion now planned for ÖйúAƬ is modest; 15,000 extra places next year will barely keep pace with rising numbers of 18 to 20-year-olds.

Securing a wider social mix of students in a limited envelope is harder. More of one sort of student can only mean fewer of another. The professional classes have long understood the value of a university education and will resent their (often expensively educated) children being bumped by less qualified students from less affluent postal districts. That resentment will be the sharper because the government, rightly, wants not just to improve access overall but to widen the social mix in all universities. The presence of research produces a pecking order well known to ambitious parents and students.

The problem is not with well-qualified students. They get into the universities of their choice anyway. Indeed, evidence for the superior performance of comprehensive pupils compared to equivalently qualified privately educated students, is probably strong enough to influence admissions staff in their favour in just the way independent school heads fear. But favouring state schools does not guarantee a broader social mix: the middle class go to state schools too.

ÖйúAƬ

The big challenge for the government is to crack the underlying problem: not enough young people from less advantaged social groups want to go to university badly enough to get qualified to do so. Oxford is right to point to schools as the source of present inequalities (page 5).

That the government is serious about wider participation, and is determined to tackle the cause is shown by this week's welcome extra money for further education. Raising participation and achievement in the colleges, which are best placed to attract less well-off students, will certainly help. But it is doubtful if it will be enough while the government continues to avoid tackling that hoary old chestnut, terminal school qualifications. These should be redesigned so that all young people, whether or not they intend to study further, run in the same race. Talk of parity of esteem between different qualifications and parallel pathways is self-deluding and leads to invidious distinctions between applicants.

ÖйúAƬ

The other urgent issue is financial support for poorer students. The move to pilot grants for 16 to 19-year-olds is welcome. If it works it should rapidly be implemented nationally. In due course the government may also need to revisit its decision to abolish grants for poorer ÖйúAƬ students.

By the same token, if, as David Albury says (right), ÖйúAƬ is serious about helping, it should engage more actively in reviewing present arrangements. Universities might, for example, consider some form of aptitude testing under the collective control of institutions to run alongside reformed school-leaving qualifications. This could help spot where poor performance is the result of bad schooling.

Making admission to ÖйúAƬ a matter of talent independent of background is an ambition most of us share. But it is doubtful if the ragbag of sticks and carrots now suggested can do more than alter things at the margin. The key to real change must lie not in bullying or bribing institutions but in inspiring and enabling students.