With this year’s intake at Blackburn College, principal Ian Clinton has begun what amounts to an extraordinary #163;1.75 million bet on his students’ fortunes at A-level. He has offered them a guarantee: attend regularly and carry out the assignments his teachers set, and the college promises that you will not fail. If any of the 350 students have no A-level passes at the end, Mr Clinton will hand them #163;5,000.

But if this is a gamble, then Mr Clinton is more like the casino than the sucker at the table. Blackburn College has a 99 per cent pass rate for its A-level courses, making it incredibly unlikely that a student taking the standard three subjects will fail them all. There are caveats, too: students must attend and attempt all their exams, must have a 95 per cent attendance record in class, and must have met minimum standards for their work handed in before the exam.

If anything, he is betting not on whether his students pass or fail - it is the college’s job to assess whether students are likely to succeed and to support them along the way - but on whether they find the idea of failure sufficiently off-putting that none of them will decide to throw their exams, pocket the #163;5,000 and put their knowledge to work later in resits. The college argues that students want to get good jobs and go to good universities more than they want cash in the short term, and says that they would not want to jeopardise their future with failure.

“I think that young people are not going to throw away two years of their life at Blackburn College purely to get #163;5,000. I don’t think many of them are fooled by the idea that #163;5,000 will set them up for life,” Mr Clinton says.

Instead, the guarantee is about expressing the college’s confidence in itself in the hope that it will rub off on students. “What parents and students seem to recognise is that we are serious about our students being successful. They seem to trust what we are about,” he says.

That vote of confidence seems to have struck a chord with students. At a time when the numbers of school leavers entering college has dipped slightly for the first time in nearly two decades, according to a survey by the Association of Colleges, Blackburn has seen the number of enrolments soar by a remarkable 43 per cent among A-level students this year. It highlights the guarantee as part of its package of measures for fighting back against the cuts to the education maintenance allowance.

Outside the college, however, the offer provoked an angry reaction. It has been called a “charter for dunces” and a “crackpot idea” for “rewarding failure”. As one comment put it: “Only a moron could come up with such a daft idea. I do not want to hear any more moans from the teaching profession about cuts in the education budget as they obviously have more money than sense to come up with something so ridiculous.”

Two themes emerged: that failure should not be cushioned, and that it was all too rare anyway. “This means that Blackburn College is going to be very strict on those signing up, or that the exams have been dumbed down so far ... Make your own mind up!”

The English exam system began by presupposing that a large number even of the elite group that took them needed to fail. Subsequent attempts to create grading systems that encompass wide ability ranges and acknowledge achievement at all levels have failed to win public confidence. The idea of the necessity of failure has refused to die.

A-levels were designed in 1951 so that about 10 per cent of candidates would always get failing grade Fs, with another 20 per cent only considered good enough to be awarded an O-level - effectively another failing grade. By 1987, fixed criteria for passing grades replaced the zero-sum game, where one candidate’s success meant another’s failure, and pass rates began their upward ascent. These reforms were in the same spirit as the introduction of GCSEs, which aimed to encompass the whole range of ability among students in England.

The philosophy was to tailor the assessment to the range of student ability, and it was now compatible with the idea of everyone passing. “Something that has a 100 per cent pass rate can be caused by people making sure they put the right people on the right programme,” says Tina Isaacs of London University’s Institute of Education.

But the desire to identify failures persisted, and was helped by the fact that the GCSE C grade was designed to equate to an O-level pass. “Everyone knew there was a difference between achieving an A to C and a D to G. It just carried through from the old O-level system,” Dr Isaacs says.

Although critics argue that exam failure has been all but abolished, in one sense the elimination of failure is a very recent phenomenon. It was not until 2006 that the absolute number of A-level exam entrants who failed dipped below 1951 levels, at about 28,000. This year, about 18,000 entries failed. If we are beginning to abolish exam failure, what effect will it have on students? Will it mean the end of a source of needless misery, or will we be losing a priceless motivator that teaches young people important life lessons?

Toby Young - journalist, founder of the West London Free School and son of social and education reformer Michael Young - was one of those to criticise the pass rate this year when he announced that it was “virtually impossible to fail” A-levels. While the standards debate often focuses on the distribution of the top grades, there is a strong feeling that, without significant numbers of students failing, the system must be a sham. “If the pass rate is so high, it robs an A-level pass of any value,” he said. “We should reclassify anything below a C as a fail.”

For Mr Young - who parlayed his O-level failure into a place at Oxford, via retakes and an incident where he missed the grades for his A-level offer but was apparently awarded a place in error - the experience of failure is beneficial in itself. Failing upwards, he once called it, cheerily recounting the nine national newspapers and magazines he had been sacked from, including one before he had written a word.

“The main shortcoming of the ‘all must win prizes’ ethos that prevails in the state education sector is that it leaves children unprepared for failure,” he says. “I think teaching children how to fail should be an essential part of their education. I failed all my O-levels bar one and it taught me how to pick myself up, dust myself off and start all over again. I retook, passed, went on to do three A-levels and got into Oxford. If I hadn’t failed my O-levels I probably would have ended up at Bradford Poly.”

It should be said that the desirability of failure does not necessarily apply to the students who this year began their secondary education at the West London Free School.

“If the national pass rate is 98 per cent, I’d expect to see a 100 per cent pass rate at the West London Free School. In a healthy system I think at least half of A-level candidates should fail, though I’d still expect the West London Free School to achieve a 100 per cent pass rate,” he says.

He is not unique in that: Mr Clinton also says he had concerns about the high national pass rate, despite his pride in his college exceeding it.

The idea that repeated experience of failure can be the engine of improvement has a long tradition among high achievers from literature to science. “Ever tried. Ever failed. No matter. Try again. Fail again. Fail better,” wrote Samuel Beckett. The physicist Richard Feynman characterised getting things wrong as the whole function of science: “We are trying to prove ourselves wrong as quickly as possible, because only in that way can we find progress.”

New research lends support to these insights. Psychologists at California University, Los Angeles, in 2009 decided to test whether the attempt and failure to recall answers in a near-impossible test made it more or less likely that the next time they would get it right. It is an idea that has useful implications for teachers: do students learn more if given tests they are likely to pass or if they are stretched, even to the point where they fail?

Students were required to learn pairs of loosely related words, such as “whale” and “mammal”, which they would have only a 5 per cent chance of guessing. One group was told the first word and given eight seconds to guess its pair. The odds meant they almost always got it wrong. After being given the right answer, they then had five seconds to memorise it.

Meanwhile, a control group crammed for the test, with 13 seconds to study and memorise each pair. Those who had failed the first, impossible test subsequently performed 10 per cent better, even after a delay of 38 hours.

So the experiment found that people remembered things better and for longer if they were given challenging tests that they were bound to fail. The effort of trying and failing to retrieve the information helped ensure that when they were given the answer, it stuck. The researchers concluded that, instead of arranging them for students to avoid errors, tough tests could be an important key to effective learning.

There is a difference between this kind of classroom test and the public exams that mark the end of many people’s time in education, however. The researchers’ test assumed that people would try again. But are those who fail their exams spurred on to do better? Or are they disheartened?

The annals of business success are littered with the self-made men who left school without qualifications but who were determined to prove themselves, from retail billionaire Sir Philip Green to diamond magnate Laurence Graff: their lack of schooling becomes part of their myth. But they are, by definition, outliers. The experience of failure for others is very different.

Russell, who asked that his real name not be used because failure still carries its stigma, first failed the 11-plus, meaning that he did not follow his brother to grammar school, but attended the secondary modern in the northern city where he lived. “When I got there, the facilities weren’t great for a start: it was pretty derelict by comparison. So from an early age we felt pretty second class. In hindsight, quite a lot of the teachers felt second class as well,” he says.

Having been anxious to succeed at the 11-plus - too anxious, perhaps - he lost interest in school. “For five years, I really just messed around,” he says. Inevitably, he left school without qualifications, and he says the experience left his confidence low.

Although he’s unlikely ever to be a billionaire, he did turn his life around from the succession of part-time jobs and occasional petty crime that he fell into, and in this sense his story is not that different from some of the business magnates. Many of them are united by having found an alternative place to learn shortly after leaving school - in several cases, it was an apprenticeship.

For Russell, it was less formal, but had a similar effect: a low-level job with an advertising agency that had a bohemian disregard for paper qualifications and an enthusiasm for helping people get on. Eventually, they encouraged him into night school, which took him to university and to the job at a children’s charity he holds today.

One obstacle to this indicates some of the damage that failure can do: his parents tried to persuade him not to enrol at night school. “When you go down that route, your own family members start to label you,” he says. “My brother was very much the grammar school boy, who was going to be an economist or a fighter pilot, and I wasn’t going to do much. It was very much the one who achieved and the one who failed. My family tried to persuade me not to go to night school because they thought I would fail again,” he says.

Why failure turns some people into billionaire entrepreneurs in the making, some into future charity workers and leaves some damaged for life is a puzzle academics are trying to resolve. David Putwain, senior lecturer in psychology at Edge Hill University, says there is little research on how school leavers respond to the experience of failure, with most psychological research on exams focusing on pre-test stress.

His interest in the effect of testing on motivation came about after working for an A-level exam board. He has identified a group of perhaps more than 10 per cent of students for whom the intense pressures of exams cause them to underperform. Some of the students he worked with all but suggested his next topic of research when they complained that one of things they hated most was when teachers used threats of failure in exams and in later life as a tool to try to motivate them.

“They say ‘If you don’t do well in your GCSEs you’ll never get to college or university. If you fail you’ll never get a good job’, and so on,” Dr Putwain says. His surveys found it was a relatively common phenomenon, although some teachers eschewed it altogether. Now he is trying to sift through the evidence of how children respond. “It has complicated effects,” he says. “It does motivate, but it has competing effects on performance for students.”

Two terms for the characteristic of students who respond well to failure have been coined: most commonly, it’s called resilience, but Professor Andrew Martin at Sydney University has more colourfully dubbed it “buoyancy”. He and colleagues identified four characteristics of the buoyant child: confidence in their own abilities, engagement with school and learning, good relationships with teachers and, above all, low anxiety about school work.

What is proving harder is finding ways to instil these qualities in children who do not already have them. A pilot programme in 2008 aimed to use psychological techniques to promote realistic thinking, adaptive coping skills and social problem-solving in children, with the aim of increasing their resilience to the bumps and shocks of school life. It was inspired by a programme in the US, which boasted measurable success in 13 controlled trials, was popular with staff and students, but produced no long-term improvements.

That does not mean the effort was worthless: it recorded significant short-term improvement on depression, attendance, and academic attainment in English. There were lesser and inconsistent effects on anxiety and maths attainment. A year later the effects had evaporated, however. One of the problems of resilience is that it does not appear to be very resilient itself.

While the UK has begun a process of creating fewer failures in school, the US is heading in the opposite direction. After years of relying mostly on teacher assessment to determine whether students earn their high-school diploma, concerns about accountability are increasingly persuading states to adopt high-stakes tests.

The raw material for one study came from a disastrous blunder by the test administrator in Minnesota in 2000, which prompted the state education commissioner to write: “I can’t imagine a more horrible mistake that NCS could have made. And I can’t fathom anything that NCS could have done that would have caused more harm to students. I know that NCS agrees with me.”

Nearly 8,000 students had incorrectly been told they had failed a maths test that the state required for graduation. Unluckily for the testing company, the father of one of the girls who was failed was a lawyer. He uncovered the error and the eventual result was a $7 million class action lawsuit, which involved researchers tracking down the borderline students who thought they had failed, and surveying them.

What they found gives some support to those who say that the possibility of failure must be there as a motivator, and that failure itself can be a spur to future success. Because while 80 per cent of students said they felt depressed, worried or embarrassed by the experience, and half said they felt “stupid”, 80 per cent also increased the time that they studied, and half cut down on the time they spent with friends, in part-time work or scrapped summer plans for additional study time. Mostly, they were spurred on to do better.

The findings also suggested that failure had to have negative consequences to motivate students. Those in lower grades of school, who would have several years in which to retake the test anyway, showed the least emotional effects and were least likely to change their work habits.

But it’s not quite the end of the story. A small proportion of the students, about 350 in all, dropped out of school permanently, discouraged and humiliated. Some of the evidence is conflicting, but subsequent studies have suggested that poor students or those from ethnic minorities are more likely to respond to failure by dropping out.

It creates a troubling dilemma: the prospect of failure with severe consequences motivates most students, while ensuring that a minority are permanently cut off from success. The authors of the Minnesota paper compared it to the pain after surgery, which surgeons say is necessary to encourage patients to protect themselves from activity that might hurt them while they heal. “How much pain is tolerable and healthy?” they ask.

That is the question behind the complaint that it is impossible to fail exams. It is a question of weighing the benefits of one form of motivation against students who are put off studying, perhaps forever.

And that, more than a stunt offering #163;5,000 to failing students or the $7 million paid to teenagers who were wrongly informed they had failed, may be too high a price to pay.