England is one of the few developed countries which allows students to give up maths before the end of schooling. Most European countries insist that students continue until age 18 or 19.

This explains why we are at the bottom of the league tables for adult numeracy, with around 15 million adults judged to be effectively innumerate. It seems unlikely that this situation will improve if we allow students to stop their study of maths even earlier, at 14.

Does this matter? Yes, if we don’t want to become a low-skills society exporting numerate jobs to better-qualified countries. Yes, too, if we want as a nation to live well. You do not have to know much maths to survive as an adult but, if you want to improve your quality of life, you need to make the best decisions about loans and debts, insurance, household purchases, pensions, mortgages and investments.

An understanding of some mathematical ideas such as ratio, percentages, probability and risk, growth rates, compound interest, variables, standard deviations, medians, trends and correlations will certainly leave you less at the mercy of unscrupulous sellers and advisers. One of the reasons adults come to numeracy classes is a great fear of being endlessly “ripped off” in a world where the state is increasingly reluctant to take any responsibility for care of the individual.

In higher education, maths is increasingly required not just as an entry qualification but also as a support subject. Not only scientists and engineers, but also undergraduates studying health and sports sciences, psychology, medicine, geography, economics, business studies, information and communications technology and design are likely to find themselves studying subsidiary maths or statistics.

One university reckoned that this amounted to two-thirds of its students.

With an intended 50 per cent of school leavers entering degree courses and others joining later as mature students, more than 40 per cent of the age-group could end up studying maths in HE. What if many of these students had given it up at 14?

There are no prizes for guessing which groups of students would elect to stop maths early - the bias is likely to be towards girls and less-advantaged students.

This would be a serious step towards reducing equity since mathematical qualifications will still serve as gatekeepers for many courses and careers. Surely, also, it is the students who have not done well in the subject pre-16 who are most in need of continuing their study.

Mathematical concepts take time to learn to a level where you can use them flexibly. Stopping too soon will remove from many people the chance of getting to grips with key ideas, and will deny them the confidence which comes with greater exposure and opportunities for application.

Ideally, of course, maths would be embedded in other subjects. However, this rarely works. Teachers of vocational areas and of other school subjects already feel overburdened by the content to be covered in their main subject and are rarely confident in teaching the underlying maths. So it tends either to get omitted or to be taught as an isolated, soon-to-be forgotten set of rules.

Certainly students, and many teachers, are bored by the subject as it is taught at present and find it difficult to appreciate its uses or fascination. However, as the Smith report points out, the remedy is to improve the forms of assessment, the syllabuses and the quality of the teaching rather than giving up.

The research literature demonstrates that it is quite possible to teach maths in an interesting way, and we have a lot of knowledge about how to do it. But it cannot be done in a world dominated by tests and league tables, over-full and outdated content, poor-quality teaching materials rushed out untrialled for changing syllabuses, and under-educated teachers.

With the implementation of the Tomlinson and Smith reports, we have an opportunity to change post-14 maths radically for the better. Getting it right is the real challenge for the new maths tsar.

Margaret Brown is a professor of mathematics education at King’s College, London