Sleep expert Dr Paul Kelley sounds a bit tired. He’s a morning type, which means he wakes at 6am after about eight hours of sleep. It’s lunchtime when we speak; he’s been up almost seven hours and he may seem weary because he’s being pursued by the world’s media who all want to speak to him about sleep – and the lack of it.

Last week, Kelley, American-born but now based in Tyneside, gave a thought-provoking speech to the British science festival in which he called on schools to let teenagers lie in and suggested that lessons should be put back to as late as 11am to address a crisis in sleep deprivation among young people.

Teenagers, it seems, have different circadian rhythms, which govern the cycle of sleep and wakefulness. When a middle-aged parent’s body clock is saying “Go to bed, it’s time to sleep”, their teenager’s is saying, “No way, I’m wide awake.” When the parent thinks it’s time to get up to go to school, the teenager’s body clock is still out for the count.

While we all thought our bed-bound teens were just being lazy or uncooperative, it turns out that their behaviour is being driven by their biology. According to neuroscientists, teenagers’ circadian rhythms typically lag a couple of hours or more behind those of adults, so current school start times mean that they are forced to wake too early and are trying to concentrate on school work when they should still be asleep.

These changes mean they are not tired at bedtime, they are exhausted in the morning and if they are forced to conform to an adult timetable, they end up more sleep-deprived than a junior doctor on a 24-hour shift, impairing physiological, metabolic and psychological health.

Kelley estimates that young people in the UK are losing 10 hours’ sleep a week on average. Instead of making teenagers fit into a conventional 9am-5pm day, he says 16-year-olds should be allowed to start school at 10am and 18-year-olds as late as 11am. Not only will GCSE results improve, he says confidently, but also teenage moods will lift, physical and mental health will benefit and family life will be more fun.

“At the age of 10 you get up and go to school and it fits in with our nine-to-five lifestyle,” explains Kelley. “When you are about 55 you also settle into the same pattern. But in between it changes a huge amount and, depending on your age, you really need to be starting around three hours later, which is entirely natural.”

I try this theory on my 15-year-old son, who initially rather likes the idea of rolling out of bed a couple of hours later. “Yes, I’d be a fan of that,” he says, then pauses. “So would the day end later?” Yes, it’s all just pushed forward. He considers. “Ah, I’m not sure about that,” he says. “You wouldn’t have time to do anything in the evening.”

Kelley, 67, a former headmaster and a father of four children, has plenty of experience of sleep deprivation. “Welcome to my world,” he says. “We’ve got four children, all five years apart. We’ve had this for 20 years.”

As a result, Kelley, who now works as a research associate at Oxford University’s Sleep & Circadian Neuroscience Institute, has long been fascinated by the growing body of research investigating sleep and its impact on learning. He comes from an education background; he worked closely with Tony Blair’s education team on its Building Schools for the Future programme to rebuild secondary schools and had a particular interest in the impact of natural light on learning.

When he became headteacher at Monkseaton High School in North Tyneside, he decided to test what he’d read about teenage sleep patterns and masterminded a school-wide pilot study looking into the effect of a delayed start to the school day. First he had to get the teachers, governors, parents and children on board, then he put the timetable back to a 10am start for all year groups.

The results were striking. Pupils’ concentration improved, days off sick decreased and the atmosphere in school changed radically. Students were nicer to each other and more pleasant to their teachers; at home, family life improved with bemused but happy parents reporting that their children were “less moody, less obnoxious”.

What really mattered, however, were exam results. “The thing about running a school is that the bottom line is better performance,” says Kelley. The late start paid off, resoundingly. At the end of the first year of the experiment, the percentage of children achieving the government benchmark of five GCSEs at A*-C including English and maths increased by 19%; the following year it went up by 18%.

“We had this very big GCSE rise. We had a drop in absence. But the best thing was the mood – it was bliss,” recalls Kelley. The two-year experiment came to an end, Kelley moved on and Monkseaton, under new leadership, resumed the conventional school timetable.

Now, however, a much bigger project is under way to further investigate the impact of a later school start on educational achievement. The £70,000 study, which is part of a wider research programme looking at the ways in which neuroscience can improve learning and exam results, is recruiting more than 100 schools to experiment with different start times. In the coming months and years, there should be thousands of happy, well-rested teenagers out there, thanks to Dr Kelley and his Oxford colleagues.

“It’s a win-win for everybody,” says Kelley. “It’s a win for parents, it’s a win for the child and it’s a win for schools and universities because they will do better.”

How practical it will prove remains to be seen. Last week, following excited coverage of Dr Kelley’s ideas, one Oxford head teacher consulted parents about a possible later start to the day. “We asked parents and pupils about doing it,” said Sue Croft, principal of Oxford Spires Academy, “but we never even trialled it because it just seemed too impractical. I just think it would have been too hard for parents to organise.”

This week, other research into teenage sleep patterns by the Wales Institute of Social and Economic Research, Data & Methods warned against meddling with school start times. “Having a regular morning routine may actually prove to be a very important feature in helping adolescents concentrate and enjoy their learning, something that may actually be undermined by changes to the school day,” the paper concluded.

As a teenager, Paul Kelley says he was sleep deprived like everyone else, though he’s always been “a bit of a morning type”. His father was the celebrated mathematician John L Kelley who had tea with Einstein and worked in topology. Kelley also went to the University of California, Berkeley at the tender age of 15 to study physics. He has now had four teenagers of his own who have all demonstrated the usual teen tendencies. His youngest daughter is 17 and still at home. “She’s not at her best at 9am,” he says.

Modern culture – the world of work, business and education – favours the early riser. Benjamin Franklin’s famous saying, “Early to bed, early to rise / Makes a man healthy, wealthy and wise”, still holds sway. Not so, says Kelley – at least not for our teenagers. Shhh! Be quiet. Let them sleep.