In 1858 the University of London introduced advanced, faculty-based matriculation examinations, which were the basis of today's A-level system. A-levels were developed as university entrance examinations, not as a stage of education in their own right. Thus it has long been recognised that A- levels may not provide a suitable curriculum for students who fail to go on to higher education. However, A-levels are undoubtedly rigorous and increasingly popular.
The proportion of 17-year-olds obtaining two or more A-level passes increased from 12 per cent in 1975 to just over 20 per cent in the mid-Nineties. The rigor and specialisation of A-levels have also ensured the continuation of three-year, rather than four-year degrees.
The disadvantage of A-levels are that students may complete the course, fail the examination and have nothing to show for two years' work. A-levels therefore represent a significant academic hurdle, particularly for the less able, and have been held responsible for the high (but falling) UK dropout rate at 16. Furthermore, students choose their own subjects at A-level and not all necessarily reach the same standard in key skills, such as mathematics and English.
We tested whether an individual's A-level subjects have a differing effect on labour market outcomes, depending on their final schooling level.
There is clear evidence of a large, positive return to mathematics A- level, even controlling for previous ability and further study at the graduate and postgraduate level.
This result is more powerful than previous research, which has indicated only a return to basic numeracy. A-level mathematics is obviously greatly valued by firms, even when individuals reach the age of 33, and after taking into account their personal characteristics, general education level, innate ability, performance at degree level and work history. The skills provided by A-level mathematics are clearly correlated with workers' future productivity, justifying employers' demands for an increase in the supply of these skills.
A possible explanation for this result is that the mathematics skills learnt at A-level, such as logical thinking, problem-solving and statistical analysis, may be closer to those used in the workplace itself than the skills developed in other A-level subjects. For example, we found no evidence of a positive return to language A-levels.
This is perhaps because these A-levels do not in fact provide the language skills required by employers, such as report-writing and verbal communication skills. Equally, we found no evidence of an additional return to A-level scientific skills.
Our evidence supports the view that more students should be encouraged to acquire advanced mathematics skills at age 16-19. The introduction of the proposed "key skills" course may be one way to do this, if the course is specifically designed to develop advanced mathematical skills and is targeted at those who do not intend to take mathematics A-level. However, simply adding mathematics to the 16-19 curriculum may not improve labour market outcomes if the academic standard of the course is not sufficiently rigorous.
Since some students drop mathematics at A-level as it is not one of their preferred three subjects, broadening the A-level curriculum may be an alternative way to encourage more students to study mathematics at 16- 19.
If most schools and colleges successfully encourage their A-level students to take five subjects in their first year of sixth form study, more young people may opt to take mathematics at AS level.
These reforms aside, the Qualifications and Curriculum Authority may have developed the most useful way to encourage greater take-up of mathematics at age 16-19. They are piloting 12 different mathematics units, to be made available to A-level and GNVQ pupils in September 2000.Reuse content