It's an anecdote the prime minister still tells today. In January 2012, David Cameron invited all the secondary headteachers from his constituency in "leafy west Oxfordshire" to 10 Downing Street. He sat them around the green baize of the Cabinet table and introduced their counterpart from an inner-city London comprehensive. The prime minister pointed out that, despite having "many children with multiple disadvantages", she had achieved better GCSE results than some of their schools.
Still digesting this uncomfortable news, the Oxfordshire headteachers were taken on a tour of her secondary school. Despite a little initial reluctance, Cameron says that some of his visitors have since enthused to him about the "fascinating ideas" they picked up.
It is an exercise that many educationalists and political leaders would now like to repeat on a national scale. The amazing improvements gathering pace in London's state schools for a decade have become impossible to ignore. As the prime minister has noted, if "four out of five schools in Surrey and Oxfordshire are doing worse than two state schools in relatively deprived parts of inner London.that must be a wake-up call".
The question is: can the capital's success be replicated in the rest of the country and, if so, how? To work that out, you need to understand how it happened in the first place, which is easier said than done.
A major report released today argues that London's success is largely down to policy, which could be good news for other regions hoping to match its performance. But there are suspicions that wider changes in the city, which go far beyond education, are playing a big role in raising school standards. The fear is that these general factors will be harder to reproduce elsewhere.
It is a complex picture that could involve everything from deliberate top-down education initiatives to gentrification, teachers' ambitious spouses, bottom-up collaboration, a sustained economic boom, optimism and high expectations, speed dating and coffee shops.
What is quite clear is that something special has happened. Chris Husbands, director of the University of London's Institute of Education, believes that the capital's state school system has achieved results that make it the "international education success story of the past 10 years".
Shanghai, Singapore and Hong Kong might have something to say about that if results from Pisa (the Programme for International Student Assessment), in which they top the table, are to be believed. Interestingly though, all three are, like London, cities that stand out from their immediate surroundings as global economic and education powerhouses.
London's figures (see panel, page 26) speak for themselves and, although the data does not yet exist to prove it one way or the other, they have led many to question whether the city has the best school system in Europe.
What is known is that London now consistently exceeds the national average on the benchmark of five GCSEs at grades A*-C including English and maths. That is an impressive feat when you consider the capital's historical record. But it is nothing short of remarkable when you reflect that last year nearly a quarter (23 per cent) of London secondary students qualified for free school meals, compared with 14 per cent in the rest of England.
As Sam Freedman, research director at Teach First (the teacher training scheme that many believe may be responsible for much of London's success), has pointed out, the GCSE benchmark is merely a threshold. It does not do justice to the greatest strength of London's schools - how well they do with poorer pupils.
The London Challenge
There is a far lower correlation between pupil deprivation and low GCSE results in London than any other region in England. Freedman has highlighted the discrepancy by comparing schools in different parts of the country that have enough disadvantaged pupils to be eligible for Teach First (at least half must be in the lowest 30 per cent of the government's Income Deprivation Affecting Children Index).
Results in 2012 revealed that the average gap between children attending such schools in inner London and those from similar schools in the two worst-performing regions (the South East and the South West) was the difference between each student achieving six D grades at GCSE and six B grades.
Last year's GCSE results revealed that those inner London schools had pulled even further away from some regions: the gap between them and their Yorkshire and Humberside equivalents was the difference between eight Ds and eight Bs.
The more cynical might wonder whether this chasm could be explained by some kind of jiggery-pokery or league table "gaming". The answer appears to be no. The GCSE point scores used in the comparisons above strip out all "equivalent" qualifications. Moreover, a high-profile report released today, Lessons from London Schools: investigating the success, states that "even controlling for `gaming', London's pupils are exceeding expectations".
"The more you look into it, the more [London's] success is apparent," says report co-author Loic Menzies. "That stands up when you look at Ofsted gradings as well." Data showing the proportion of students going on to further and higher education tells a similar story, he adds.
The report argues: "The improvement cannot be explained in terms of the advantages that London has over the rest of England." Some disagree with that clear-cut view. But there is a general consensus that certain government initiatives and interventions have at least contributed to the gains.
Despite the prime minister's evident pride in London's schools, most of the policies credited in today's report can be traced back to before the coalition government came into office. Indeed, the study pays tribute to the "pioneering" use of educational data started by the Inner London Education Authority, which was abolished back in 1990.
The report, produced jointly by the CfBT Education Trust and the Centre for London thinktank, also identifies more recent developments. Extra school funding, better teacher recruitment and improved school buildings are cited as three important "enabling" factors for London's success. It also credits a new culture of high expectations in many of the capital's schools, and of "no excuses" for underperformance.
But it is Teach First, improved support from local authorities, the academies scheme and the London Challenge that the report describes as the four "key" interventions.
The latter is probably the policy that many people would consider to be responsible for the turnaround, as the timing is a near-exact fit. When the London Challenge began in 2003, the capital had the lowest proportion of students achieving the GCSE benchmark out of England's nine regions. By 2006, it was in fourth place and by 2011, when the coalition ended the scheme, London was comfortably topping the table.
The London Challenge has acquired a reputation for being that rare beast, a government school improvement scheme that worked. However, at its official launch the signs did not seem particularly good. Glossy brochures were printed and journalists invited to a ministerial speech. But to some it appeared to be little more than reheated academy announcements cobbled together with gimmicky-sounding schemes about helping poorer children to go to the theatre.
Time has proved such cynicism wide of the mark. In retrospect, what appears to have been the most powerful aspect of the London Challenge is the change in mindset it engendered.
"In the early 1990s, there was fierce competition between schools," says Dame Joan McVittie, headteacher of Woodside High in North London. "That was the coin that London Challenge flipped, with school-to-school support and schools working together. And that is why [London] is still improving, because it broke down the barriers. Before that, you pulled up your drawbridge and you didn't help anybody."
A wild-haired `tsar'
At the centre of the new collaborative culture was Sir Tim Brighouse, London's first schools "tsar". The wild-haired, well-respected veteran of Birmingham's education authority set the tone with his visits to London schools. He treated teachers as equals, wrote them personal letters and talked about their work using words such as "exceptional", "amazing" and "brilliant". Above all, Sir Tim (or Professor Brighouse, as he was known back then) listened.
His idea was that the London Challenge was something done with schools rather than to them. And his conciliatory approach meant that schools took notice of the call to work together. "Headteachers told us that you couldn't say no to Tim Brighouse," Menzies says. "One said: `You walk over glass for people like that.' "
The geographical density of London and good transport links also helped the collaboration, as did the language used. Sir Tim decided to refer to schools needing the most help as "Keys to Success" schools, rejecting terms such as "poor" and "underperforming". Today's report reveals that the London Challenge became such a positive brand with schools that one borough's director of children's services applied the title to things that they now admit had nothing to do with the scheme.
The language lesson went unheeded, however, when the Labour government tried to apply the initiative countrywide through the much-resented National Challenge programme. It came with pound;400 million to help underperforming schools, but the use of words such as "failure" in relation to an easily identifiable list of secondaries threatened with closure ensured that there would be no repeat of the London feel-good factor.
A better clue as to whether the London Challenge could be applied to other parts of the country came when the similar Greater Manchester and Black Country Challenges were launched in 2008. Freedman wrote last year that the two regional school-improvement schemes "weren't nearly as successful" as the London version, "perhaps because the necessary concentration of existing [teaching] expertise wasn't there to make it work".
He has since, like Menzies, acknowledged that although the signs were positive, three years wasn't long enough to make the same difference. But Freedman, a former adviser to England's education secretary Michael Gove, has also pointed out that, while plenty of school leaders involved with the London Challenge believe that it worked, "there is little in the way of hard statistical evidence" to prove it.
He agrees that the capital has been helped by having more than its fair share of the "best" academy sponsors. But unlike Menzies and the authors of today's report, he thinks that its phenomenal success could also be down to factors beyond the control of education policymakers, such as the city's greater prosperity and the accompanying gentrification.
Freedman has noted that the boroughs that have seen some of the biggest changes in socio-economic status over the past decade - such as Southwark, Tower Hamlets and Hackney - have also seen the greatest gains in school performance. He even came across a study that plotted London's good coffee shops on a map to show how they corresponded to gentrification. When the areas seemed familiar to Freedman, he quickly realised it was because the coffee shops roughly matched the distribution of the capital's best state schools.
But this apparent link between superior cappuccinos and education has not been taken up by the Lessons from London Schools report. It rules out the gentrification, stating: "The improvement in educational performance results from the better performance of students, including poor students, and not the displacement of deprived students by more advantaged groups."
Freedman acknowledges that secondaries in the gentrifying areas "haven't suddenly filled with yuppies' kids". But his argument is that the poorer pupils who remain in these schools can have their aspirations raised by the prosperity growing around them.
"There are lots of poor families in Tower Hamlets but they can see Canary Wharf and they can see the Olympic Park and they are two stops away from the centre of London," he explains. "That is very different from being in Doncaster or Great Yarmouth, which are a long way from anything. There is a sense of being in a place that has a lot of possibility."
Whether the poorest children in East London feel any real connection with the centres of global finance that tower over their homes and whether they actually make the short journey into the West End is open to question. The theory also fails to explain why Haringey has had roughly the same rise in GCSE results as Tower Hamlets but a fraction of the gentrification. Meanwhile, the residents of Doncaster may not feel that they are living in the wasteland that Freedman perceives.
Nevertheless, the general idea is an attractive one for those who believe that schools and their students cannot be seen in isolation. Can it really be coincidence that at exactly the same time as London's economic performance is lifting the capital way above the rest of England, its schools are performing the same feat in education?
Today's report notes that London's state schools were doing far worse in the 1980s and 1990s, when London was also booming. But, Freedman counters, "there was a quite narrow boom in the 1980s, very much focused around finance. What is happening now is much more widespread and cultural - London has shifted from being a UK city to being a global city. It probably did start back then, but I think it has been a long process to get to where we are now."
One area where economics and education policy almost certainly do come together is that of teacher supply. Menzies points to improved pay in inner London and the success of Teach First in attracting bright young graduates to teach in the city's most challenging schools. The report he helped to write quotes a former London headteacher, who says: "The holes in the staffroom were being filled by very, very bright, very energetic young people instead of [.] rather demoralised supply teachers or people from abroad."
But Menzies' own experience helps to explain why London's general economic status may also have helped. He worked as a teacher in London after joining Teach First, but was brought up in Cambridge and studied at the University of Oxford. Asked why he then chose to work in the capital, Menzies admits: "It was sort of an assumed thing - you finish university and you go down to London. That is what you do."
So the natural tendency is for graduates leaving England's top universities to head to the capital. And with better pay on the table and the idea of teaching in challenging schools detoxified by Teach First, London's state schools are now also able to benefit from this talent suction pump.
But as Teach First seeks to foster similar improvement in the rest of England, it is facing the flip side of the phenomenon that it benefited from when it first set up in the capital. "We can direct people to certain parts of the country and they usually agree - that is easy," Freedman says. "Our mission now really is to help to persuade people to stay there."
Creating a buzz elsewhere
One strategy has been Teach First's decision, revealed this week, to change its eligibility criteria in order to allow more schools in more remote but educationally underperforming areas to take its trainees. That could help to create a cluster effect, making teaching in unfashionable towns more attractive to bright, energetic and ambitious young trainees.
"If you put one person into Great Yarmouth, they will really struggle," Freedman says. "But if you put in 20 people and you make it into a community and a mission around improving education there, you can build something around that."
But it is not just teachers who will need to be persuaded. Their partners are also likely to be ambitious graduates looking for jobs. And that is where London could hold the trump card over places such as Great Yarmouth or Grimsby, regardless of what policymakers do. One commentator joked that speed dating between Teach First trainees and local non-graduates might be the answer.
None of this should suggest that London's state schools have reached perfection. No evidence is available to show whether the capital is a true global education leader because, like most cities, it is not entered separately into international comparative studies. But Andreas Schleicher, the coordinator of Pisa, has warned that if London were, it would "not be anywhere close to some of the [East Asian] city states".
Moreover, in 2012, London mayor Boris Johnson's team suggested that the city's schools still suffered from a "culture of low expectations". Many suspect this was part of an attempted power grab by Johnson. If so, Chris Cook, the former Financial Times journalist whose analysis reveals just how far London education has pulled ahead, is likely to approve. He argues that the needs of the city's schools are now likely to be so different from the rest of England that it could probably benefit from having a pan-London education authority.
There is no doubt that the Department for Education has more than enough on its plate with education outside the capital. In the North East, Freedman has highlighted how Ofsted judgements on the quality of teaching and school leadership are comparatively good for primary schools, but drop off a cliff for secondaries serving the region's most deprived young people. Does he really believe that London's example can be used to turn such situations around?
"I think it is possible but I think it is very hard," he admits. "The last government put a lot of money and effort into London and other cities, and that has had a very positive effect. But when Teach First began, people would say: `London has got all these problems; you can never make it as good as the rest of the country.'
"Now that has switched and we need to change the mentality again. We now need to think: `How do we solve the problem everywhere else?' "
In 1987, less than 11 per cent of inner London students achieved five "good" O-levels, compared with 26 per cent nationally. By 2002, the gap had narrowed, with 41 per cent achieving five A*-C GCSEs, compared with 49.5 per cent nationally.
The following year, the percentage of all London students achieving the benchmark of five A*-C GCSEs or equivalent, including English and maths, improved to match the national average.
In subsequent years, London schools have gone on to better this: the capital's secondary students' attainment has outstripped improvement in other parts of the country.
The most recent GCSE results, in 2013, saw London achieve a 2.7 percentage point increase in students achieving the benchmark of five A*-C GCSEs including English and maths, compared with a national 1.7 percentage point rise.
Between 2003 and 2011, London moved from being the lowest performer of England's nine regions at GCSE to being the best.