When Ines Alvarez-Garcia finished her PhD, her heart was set on an academic career in biomedicine. She knew that getting a permanent academic position would be tough, but she was prepared to make sacrifices.

Six years and two postdoctoral positions later, the Spanish researcher found herself at a crossroads. She still loved science and her project in a University of Cambridge laboratory was producing exciting results. However, the grant that had funded her position for three years had run out and the best her boss had been able to do for her was to secure a six-month extension while the outcome of another grant application was pending.

"The outcome of the application was totally uncertain. I got scared. I didn't want to be at home looking after the children; I wanted to be working," she says.

After five months of applications to biology journals she was offered a job as a scientific editor at the journal PLoS Biology. With her contract about to expire and a decision on the grant application still a month away, she felt she had no option but to accept it.

ÁñÁ«ÊÓƵ

In the event, the lab got its grant. Alvarez-Garcia admits she spent her first year at PLoS pining for the bench, but she is now happy to be out of the laboratory rat-race.

"I was tired of looking for funding; psychologically it is very hard," she says. "I left a job that I really liked, but I now have something with a secure future."

ÁñÁ«ÊÓƵ

Although her story will sound familiar to the many researchers who find themselves in a similar position, it has not always been this way. According to Cathee Johnson Phillips, executive director of the National Postdoctoral Association in the US, postdoctoral positions once came with a virtual guarantee of tenure. However, a steady rise in the number of postdoctoral researchers means the picture is very different today.

Part of the increase is attributable to the fact that postdoctoral positions now proliferate far beyond their origins in the physical and life sciences. Today only half the posts are found in these disciplines.

According to the US National Science Foundation's Science and Engineering Indicators: 2010 report, the reasons for the swell in numbers are not well understood, however. It says that "increases in competition for tenure-track academic research jobs, collaborative research in large teams, and needs for specialised training are possible factors".

But the reality is also that senior academics have realised that postdoctoral researchers are "highly trained professionals on the cutting edge of their field who will work for relatively low wages", says Johnson Phillips.

The pattern has been replicated in many countries. In Sweden, for example, recent increases in research funding have been consciously focused on increasing the pool of PhD students and postdocs. Little money has been spent on establishing more permanent positions, according to Stefan Thor, professor of developmental biology at Link?ping University, and this has created "a lot of friction".

"It increases the junior cohort of researchers but provides no new permanent jobs for them. Finding a proper balance in the academic career pyramid is important, with a reasonable cut-off at each step," says Thor.

For those hoping to progress to a more stable academic career, the figures make for depressing reading. The NSF estimates that only 26 per cent of recent PhD recipients in the US will secure a tenure-track position. UK postdocs appear to have even more reason for pessimism: according to the Royal Society's 2010 report The Scientific Century: Securing our Future Prosperity, 30 per cent of science PhD graduates go on to postdoctoral positions, but only about 12 per cent of those attain permanent research jobs.

The situation has been thrown into relief even more starkly by the global economic downturn, which, particularly in the UK and the US, has made many universities reluctant to take on new permanent staff.

ÁñÁ«ÊÓƵ

Another factor contributing to the scarcity of permanent posts is the unwillingness of senior academics to retire and free up positions, according to some. Bob Williamson, secretary for science policy at the Australian Academy of Science, is 73, but he only recently retired as director of the University of Melbourne's Murdoch Children's Research Institute. He says his decision was driven in part by a "genuine belief that older academics should make way for younger ones", but adds that older academics are often put off from retiring by pension schemes that "force you to work either full time or not at all".

"We have to find a way to allow people to give up their administrative responsibilities but continue to do useful work intellectually, perhaps on a lower salary," he says.

Worryingly, there is a big mismatch between postdocs' expectations for their careers and the reality. In a recent survey of chemistry and physics postdoctoral researchers by the UK's Institute of Physics and the Royal Society of Chemistry, 65 per cent of the 800 respondents said that their most likely career destination in the next six to 10 years was "academic on permanent contract".

So what should be done? Everyone agrees that postdoctoral researchers have to be made more aware of the low statistical probability of their being able to remain in academic science long term. Whether this would make any difference to their expectations is another matter.

Jenny Rohn, an American postdoctoral researcher working in a University College London cell biology lab, thinks not. She argues that "it is human nature to think you will be the exception".

"We all thought we were going to win Nobel prizes - it is what you do when you are young," says Rohn, who is also an author and blogger and the instigator of the Science is Vital campaign that was credited with helping to protect the UK research budget.

What would make a difference, according to Williamson, would be for universities and labs to cut the number of postdoctoral positions they offer and increase the number of senior posts.

"We are probably giving too many people an opportunity to think they are going to make it in the system," he says, adding that the challenge of remaining in academic science can become a test of endurance rather than a test of scientific ability. "When you have 50 to 100 qualified people applying for every senior post, you have to ask whether the balance is correct."

However, he is not confident that any government under pressure to create jobs would support reducing postdoctoral researcher numbers.

James Lightbourne, director of the division of graduate education at the NSF, is also wary of restricting junior researchers' options. "Some people would do anything to stay in science and who is to say they shouldn't be able to do that?" he asks.

Others point out that it is impossible to predict how many scientists will be needed in the future.

For Johnson Phillips, the key is for research organisations and principal investigators to stop seeing their postdoctoral researchers merely as "extra pairs of hands" and to take responsibility for providing guidance for careers beyond academic science as well as within it.

But postdoctoral researchers themselves are often willing partners with their supervisors in the "conspiracy of silence" about their future destinations, observes Richard Joyner, emeritus professor of physical chemistry at Nottingham Trent University.

"Both stick their heads in the sand and hope a suitable academic position will come up - and then throw up their hands in horror when the money runs out and the postdoc has to be made redundant," says Joyner, a member of the Research Careers Initiative panel, which examined the career development of UK researchers in the early 2000s. The solution, in his view, is an annual meeting between the postdoctoral researcher, his or her supervisor and a careers adviser, at which an "honest assessment" is made of where the researcher's career is going.

Such practices, however, appear to remain the exception rather than the rule. More than half of the respondents to the IoP/RSC survey had never been appraised during their postdoctoral careers, and only 45 per cent had received career advice during their current contract. The vast majority of that advice came from academic staff, leaving just 23 per cent of respondents feeling that they had a good awareness of career options outside the academy.

Realistically, some principal investigators are simply not cut out for, or are too busy to take on, mentoring roles. But where this is the case, they should become sufficiently self-aware to pass on their postdoctoral researchers to someone who is capable of mentoring them, Johnson Phillips argues.

She also urges researchers to proactively attach themselves to more than one mentor, and says that principal investigators should make it clear to their postdoctoral researchers that they are welcome to take time out to attend career development seminars organised by graduate schools.

In addition, another cultural shift is needed, according to Sir Mark Walport, director of the Wellcome Trust: the science community must cease to regard those who leave academic science as failures.

"If you do history at university, people don't assume you will be a historian," Sir Mark told a recent discussion organised by the Campaign for Science and Engineering. "History is viewed as a pluripotent stem cell that can lead to many careers, but science is not viewed in the same way."

David Price, vice-provost for research at University College London, agrees. He thinks that researchers should be encouraged to move back and forth between universities and industry and that academic research careers need "well-defined exit and re-entry points". Moreover, he welcomes the fact that not all of those on short-term contracts will immediately secure a permanent academic position.

"There will always be an oversupply of people competing for positions at each level of the academic pyramid. But this is a good thing - not only for science but also for the economy, since doctoral and postdoctoral training are useful for industry, business and management jobs too," he says.

ÁñÁ«ÊÓƵ

The sense of career disempowerment experienced by many postdoctoral researchers is exacerbated by the fact that - as in Alvarez-Garcia's case - their salaries are often funded by their supervisor's grants, which rarely last beyond three years. This means they are often on short-term contracts and not treated as full members of academic staff.

UCL is among a small group of universities that has attempted to address the problem by moving its postdoctoral researchers on to open-ended contracts. Price thinks this makes a "cultural" difference and arguably offers "some increased security and more flexibility in pursuing other funding sources".

However, the move has not won over Jane Thompson, national industrial relations official at the University and College Union. She says that open-ended contracts do not feel very different to the status quo because postdoctoral researchers are still made redundant when grants run out. She questions why researchers should be treated differently from other university staff, such as lecturers and administrators, and argues that universities should act like businesses and divide funded projects between existing research staff.

Those who support the status quo often say that the existing system serves to bring in young blood. But this does not make sense to Thompson. "No one says that unless we move human resources managers on every three years we won't get young blood in the system," she points out.

UCL's Rohn has one year of postdoctoral funding left and has no idea what happens next - particularly given that she is already 43. "They say they aren't ageist but it gets increasingly hard to get funding," she says.

She is sure that principal investigators "care deeply" about their postdoctoral researchers, but believes it is "cruel and unacceptable" for universities to be "complicit in a system that nurtures you only to fuel a research machine that seems to care very little for your fate once your contract has run its course".

Rohn also argues that the high turnover of postdoctoral researchers is bad for science, not least because principal investigators' huge administrative workload means that most bench work - and significant amounts of student supervision - is done by postdoctoral researchers. They build up an in-depth knowledge of the latest experimental techniques and this means that it is a huge loss to a lab when one leaves.

Training so many postdoctoral researchers for so few principal investigator positions is a waste of taxpayers' money, she believes. "If an average tenured lab head produces 40 trainees in his career lifetime but needs only one replacement, that's 39 people who have been trained for many years at great public expense only to have to leave science."

Moreover, the chemical reagents used by researchers in bioscience are very expensive and, unlike lab equipment, cannot usually be passed on to the next promising young researcher because they have such specific purposes.

As Link?ping University's Thor puts it, funding researchers for a few years and then forcing them to leave science is "akin to every five years tearing up one freeway to build a new one right next to it".

One idea, set out by Rohn recently in the journal Nature, is that research organisations should recruit at least one permanent postdoctoral-level researcher per lab. She admits that such posts would carry higher salaries, which would leave less funding - and therefore fewer positions - for more junior research staff. But she argues that smaller, more experienced research teams would be more productive than labs full of PhD students, who "you watch breaking glassware for three years".

Such positions are relatively common in some European countries, such as France, where researchers are often employed by the government, and Sweden, where researchers are typically permitted to remain in postdoctoral positions for only four years.

Thor has between five and 10 such positions in his department of 400 people. "If things haven't gone that well for you, or if you just don't want to become a PI, this type of position can be a very good solution for everybody," he says. "They are quite valuable. And in general they seem pleased with where they are at."

But the French system has been accused of a lack of vitality as a result of too many permanent positions. In contrast, in Australian universities, even senior appointments often last for just five to 10 years.

For the Australian Academy of Science's Williamson, a balance is required. "Labs can get too comfortable as well as not comfortable enough," he says.

While some worry that creating permanent postdoctoral researcher positions would risk keeping scientists in "second-class jobs" for their whole career, the idea appears to be gaining ground in the US.

Princeton University president Shirley Tilghman is currently chairing a National Institutes of Health panel looking at the future of the biomedical workforce. In a recent interview she mooted the idea of reducing the number of PhD students and increasing the number of "permanent employees" in labs.

Meanwhile, in the UK, David Willetts, the universities and science minister, has spoken of the many emails he receives "from frustrated 33-year-olds who can't quite see where they are supposed to be going". At a debate last week, he said his department and the Royal Society would be holding a round table meeting to "work through the issues", although he emphasised that the government would not impose changes.

If all else fails, might postdoctoral researchers take matters into their own hands by withdrawing their labour?

Rohn thinks this unlikely. "If I down tools I am harming my career. Who will martyr themselves for the next generation?" she asks.

However, the success of the University of California's 6,000 postdoctoral researchers last summer in securing an employment contract that guarantees them basic rights suggests that the mouse could roar yet.

A strike authorisation ballot was approved by 90 per cent of members of Postdoctoral Researchers Organize, a union group representing postdoctoral researchers at the University of California formed in 2008.

Although industrial action ultimately proved unnecessary, Neal Sweeney, a postdoctoral Fellow at UC Santa Cruz and a member of the negotiating team, believes that his fellow postdoctoral researchers would have acted on their threat if it had proved necessary.

"At the end of the day, postdocs are no different from any other employees. Someday, somewhere, strikes are going to happen if conditions don't improve," he says.