Today's new graduate students generally begin their studies expecting to follow the well-trodden path to academic success: earn a PhD, do a short stint as a postdoc, get a faculty position, and work feverishly toward tenure. Yet a growing body of evidence shows that many of these students will someday find themselves traveling a different road. With the explosion of biotech and the corresponding rise in patent law, science policy, and journalism, biomedical PhD graduates have more career choices than ever.
Recently an NIH working group analyzed the increasingly diverse biomedical PhD workforce, and in 2012 published a report summarizing current employment trends.1 Perhaps their most important finding is that, in comparison to the 1993 workforce, there has been a surge of graduates in non-research positions and a corresponding decline in tenured academic jobs.
Roughly a third of employed graduates in the 2000s were working outside the "traditional" research sectors of industry and academia, with a surprising fraction (~13%) in non-science-related employment (Fig. 1).2 Even those in "traditional" research sectors reported a staggering range of non-research primary work duties, which likely vary often, given the frequency of reported career changes (Fig. 2, 3).2 Collectively, these figures suggest that future graduates will require an arsenal of skills to successfully navigate the dynamic job market.
The view from the trenches
As a senior graduate student, I've seen dozens of my peers go through the process of selecting a career with considerable anxiety. Even among the best-prepared students, the experience can be more like a day of reckoning than an enlightening trip to the bountiful career buffet. Despite this problem, few studies have focused on graduate student and postdoc career preferences and perceptions.
A 2011 study of doctorate students at UCSF3 suggests that the reason for this career anxiety is a surprising mismatch between the 20th century expectations of entering students and current career outcomes. For example, over 80% of students beginning their PhD planned to pursue a research career and over 50% expected to become principal investigators in academia. These numbers shift midway through graduate training, becoming more similar to the career outcomes presented in the NIH report. Although reasons for this shift vary, many students cited negative perceptions of being a PI due to the perceived high competition for jobs and funding.
This UCSF study, if indeed representative of all biomedical trainees, makes me question whether we should reconsider our current measures of training "success." I've often heard, in seminars or casual discussions, the low unemployment rate of biomedical PhDs (<2%) and expanse of available careers cited as indicators of a healthy workforce. Yet some data suggest that employment alone may not be a sufficient indicator of workforce health, since graduates report taking jobs that are less related to their degrees (Fig. 4).4 Perhaps a more accurate measure of training success would be to ask graduates whether the job they got was the one they wanted and whether they feel a PhD was necessary for their career path.
Is there a postdoc bottleneck?
The change in graduate students' career preferences due to the perceived high competition for research jobs points to one of the central concerns of trainee scientists everywhere – the issue of supply and demand. There is arguably no group more affected by this issue than postdoctoral researchers. Unfortunately the preferences and concerns of postdocs are hard to quantify, as few articles in the career and training literature give them more than a passing reference. Even they seem to hold their own positions in little esteem; almost a third of biomedical postdoc respondents to a 2006 NSF survey indicated that they took their positions either because it is expected in their field or because they could not find other employment (Fig. 5).2 I have often heard postdocs claim that they're stuck in their position only because there are too many biomedical PhDs, creating a bottleneck for transitioning into a career.
The fact that many postdocs consider their positions to be an obligatory pit stop is perhaps the reason why recent PhD graduates who take postdoc positions report lower levels of overall job satisfaction than those who enter the workforce directly (Fig. 6).2 However, there are likely many other contributing factors. If career preparation for graduate students is a few steps behind the curve, training for postdocs is lagging by miles. Most postdocs are in the paradoxical position of being paid the salary and benefits of a trainee despite receiving very little or no formalized training. More studies on postdoc preferences and more structured plans for transitioning postdocs into permanent positions would improve the sustainability of this oft-neglected group of trainees.
Challenging the pyramid
One explanation for why career selection can be challenging is that the average student or postdoc interacts almost exclusively with academics during his/her training, creating a disproportionate emphasis on academic research. Although I'm at a university situated in a biotech haven, where attitudes toward non-academic careers are generally positive, my interaction with industry professionals has been limited to the occasional panel or seminar. For non-research careers my contact has been even less frequent. It's surprising to me that training institutions and non-academic professionals have not worked harder to establish joint training and mentoring initiatives, or to diversify the training degree programs beyond the standard 3- to 5+-year research PhD.
Training programs, and arguably all basic research, would benefit from increased diversity within academia as well. I've often joked with other students that academia is a pyramid scheme, with hoards of grad student and postdoc workers scrambling upwards to join the small number of tenured elite. The NIH report recommended breaking this paradigm by providing more funding for long-term, mid-level academic positions like staff scientists and other research associates. I'd expect that most PIs, if given the opportunity to hire permanent staff, would not hesitate to do so.
Current funding schemes, however, are barriers to this change. Perhaps uniquely among STEM fields, biomedical graduate students and postdocs are both trainees and workers, generating the data that help labs secure the funding to continue their research. As funding becomes scarce and data more precious, there is an incentive for PIs to hire more (inexpensive) trainees – regardless of the opportunities available to those trainees when training has ended. This disregard for the job market in trainee hiring decisions may explain the single factor most strongly correlated with the number of biomedical graduate students. Identified by an NIH working group: the size of the NIH budget.
This dilemma suggests that the most crucial step for creating a well-trained, sustainable workforce is to provide additional funding opportunities for a variety of research positions. The NIH working group recommended giving staff scientists a greater role on grant applications, increasing the number of pre- and postdoctoral fellowships, and decreasing the number of trainees funded by lab grants. I'm skeptical that these changes will occur at the pace that they'll be needed, and certainly not without considerable advocacy and restructuring of institutional training programs.
In the meantime, scientific organizations like the ASCB are making tremendous efforts to fill the gaps in career development. My hope is that the career development initiatives at the ASCB, and the many career services officers at institutions worldwide, will stimulate sweeping changes in funding and education to ensure a healthy and sustainable workforce for future generations.
1Due to a lack of comprehensive studies, the NIH report focused on major trends identified from a mixed bag of statistics collected by the NIH, NSF, National Research Council, and others. Their major source of data was the NSF Survey of Doctorate Recipients (SDR), a career survey distributed to ~30,000 science, engineering, and health PhD graduates roughly biennially from 2001 - 2010.
2Data from the the NIH working group report. All subsequent figures were generated from my own analysis of the public-use SDR data, which is available here
3Fuhrmann CN, Halme DG, O'Sullivan PS, and Lindstaedt B. (2011). Improving graduate education to support a branching career pipeline: recommendations based on a survey of doctoral students in basic biomedical sciences. CBE-Life Sciences Education 10,239-249.
4These data may also be a product of the effect of the 2008 recession on the overall economy; however, the NIH working group did not find a similar trend in chemistry PhD graduates during the same time period.