To judge from the latest research on adult participation in learning, the answer seems to be a big thumbs up. Figures published by the National Institute of Adult Continuing Education (TES, May 11) show a clear rise in the overall numbers of Britons who take part in organised learning. They also show that participation has risen among all social groups, and not just those who are already well educated. NIACE singled out the rise in participation among part-time workers as evidence that part, at least, of New Labour’s agenda was coming along nicely. This all looks like clear evidence of real achievement, then, and to an extent it is.

Behind the headlines, though, lurks a more blurred and worrying picture. First, the NIACE figures show that the growth in participation is highest for the middle and professional classes and lowest for people from manual working-class backgrounds. New Labour has raised the lifelong learning ladder, pulling up participation levels among all groups, but the distance between the rungs has grown wider rather than narrower. While this might be good news for skills acquisition, it points to a widening learning divide.

Second, a lot of the growth in participation is not really taking place among adults, but is the result of higher staying-on rates among young people. Data from the Labour Force Survey show a number of interesting trends in the age distribution of adult learning. In every year since 1996, the fastest rise in participation has been among the 16-19 age group, followed closely by the 20 to 24-year-olds. The proportion of all learning episodes undertaken by people aged 25 and over has steadily declined. Less than a tenth of all learning episodes in the Labour Force Survey are now undertaken by people aged 40 or more. In other words, what we are seeing may be an extension of initial education, but represents a relative decline in lifelong learning.

The Labour Force Survey sheds light on another finding in the NIACE study, which showed sharp rises in participation by part-time workers. If true for part-time employees generally, this would be a very important development indeed, as employers are traditionally thought t favour investing in training primarily for the shrinking number of core workers, neglecting the expanding periphery of flexible staff. Sadly, the LFS data confirm that most part-time workers are still less likely to participate in training than full-timers, with one large and important exception: students. Among full-time students - who now make up two-fifths of the part-time workforce - around 33 per cent take part in job-related training, compared with 20 per cent of full-time workers and 11 per cent of other, non-student part-time workers. At a guess, this training consists largely of induction courses for people whose job is a by-product of full-time study. It is an interesting phenomenon, possibly quite an important one, but not particularly good news for lifelong learning.

So how is David Blunkett’s contribution to lifelong learning to be remembered? His period in office has launched a vast fleet of initiatives to promote lifelong learning, including Learndirect and Individual Learning Accounts. He pushed his Cabinet colleagues into revolutionising the planning and financing of further education and training. Some of the dafter legacies of the Tory years - such as the artificial divide between award-bearing and non-award-bearing adult education - were dumped. But the impact of these measures will only be felt in the long term. In election year, what we can see is that the largest rise in participation in adult learning is a by-product of quite another policy decision: the changes in higher education student funding.

One message, then, seems to be that messing around with the labour market seems to influence lifelong learning more than changing the educational institutions and relying on appeals to employers’ good nature. But this is to pre-judge the longer-term impact of Blunkett’s flotilla of new measures. The next secretary of state could do worse than maintain the current policy momentum, while continuing to monitor carefully the results, and holding in reserve the capacity for legislating. In particular, he or she should consider whether a legal entitlement to paid educational leave might help to engage all those groups of learners - manual workers, the over-50s, flexible workers who are not full-time students - who otherwise are barely given a sniff of decent opportunities for learning.

It would be a great pity if the Learning Age were to entrench a socially damaging and economically wasteful gulf between the knowledge rich and the knowledge poor.

John Field is professor of lifelong learning at the University of Warwick.