Despite the fact that PhD students express less interest in academic positions as they progress through their graduate training, the number of scientists with biomedical PhDs starting in postdocs has grown dramatically. This begs the question of what the purpose of the postdoc position is, and what it should be. Currently, postdocs are being treated as cheap labor, and are not being trained for independence. This inevitably leads to them furthering their PIs agenda instead of their own, ultimately hurting their own professional success and the sustainability of the enterprise itself. Many of these issues have been highlighted in a recent Twitter thread on the postdoc workforce by Future of Research ED, Dr. Gary McDowell.
Why pursue a postdoc?
Currently, it is not clear why PhD students pursue postdoc positions and how their plans depend on individual-level factors, such as career goals or labor market perceptions. Surveys have suggested the most commonly reported reason for becoming a postdoc is the expectation for one’s career. Postdocs who had not reported prior postdoc plans were motivated by having difficulties in finding another job (especially in the biological/life sciences) and the desire to have more time before deciding on a long-term career (especially in other fields).
This is despite that, from a financial perspective, a postdoc position can be a poor investment. Salaries of those who started in a postdoc averaged $12,002 lower at 10-years post-PhD compared to those who skipped postdocs, and the salaries of non-postdocs remained significantly higher for the first 13-years post-PhD. This suggests work experience may be more valuable than years of postdoc experience, again questioning the value of the postdoc years for future career success.
What is a postdoc?
Postdocs have existed for 100 years, and few were initiated by design. The first systematic study of the postdoc, called “The Invisible University,” was published in 1969, followed by regular studies focused either on the postdoc or mentioning it as part of a larger discussion of scientific training. This study contained the idea that the postdoc position was “not consciously or intentionally undertaken by the university,” which leaves one wondering what the postdoc position was intended to be in 1969!
One idea is that the postdoc position is meant to be a training period leading to the development of an independent scientist. However, that may not reflect today’s reality. Part of how the National Postdoctoral Association defines the postdoc is a “temporary period of mentored research and/or scholarly training for the purposes of acquiring professional skills needed to pursue a career path of his or her choosing.” Therefore, this does not imply training for (only) academic careers. Moreover, training independent investigators also requires independence of thought, as well as someone who is able to “define the problem of interest and/or to choose or develop the best strategies and approaches to address that problem.”
The Invisible University
Many of the issues related to the postdoc position existing in 1969 are the same today, as are the recommendations for change, although more attention is being paid to these issues today. Some notable suggestions from the 1969 report include:
- Thinking of the postdoc as someone who is “in the process of development and not primarily as a means to accomplish other ends.” This is important given that postdocs are often used as cheap labor in order to fulfill their PIs needs and expectations.
- Ensure the postdoc “need not depend on his mentor’s sources of support to carry out his proposed research.” This goes back to the idea of building independence, and part of this is the ability of the postdoc to have financial independence from their PI, thereby allowing them to pursue their own scientific direction.
- The “number of postdoctoral opportunities available any time should be related to the number of PhDs and professional doctorate holders who can profit from the experience.” This point is critical in the context of the number of postdocs driving the enterprise, in that we should not be training more postdocs than the system can handle in order to foster their professional development and career success.
Subsequent reforms proposed to the postdoc position address a number of issues including transparency surrounding the postdoc position, work-life balance, and the instability of the postdoc position itself, and many of these factors drive researchers out of academia.
However, these reforms have not yet been sufficient to effect change. Additional factors also contribute to exacerbating these issues, including the fact that training postdocs is largely not the priority of the enterprise, and therefore postdocs are largely considered as only a pair of hands, or cheap labor at the bench. As stated previously, the agenda of the PI to keep their lab’s research going may therefore be counteractive to allowing the postdoc to develop their own direction for succeeding in their career goals.
Some of the major issues surrounding the postdoc position come from the fact that we don’t know how many postdocs there are, how much they get paid, or how to classify them based on job titles (Figure 1). Since 2016, I have been involved in studying policies related to the postdoctoral population as part of the non-profit organization Future of Research (FoR), where I now serve on the Board of Directors. Our studies seek to address some of these questions related to postdocs.
Large fluctuations in postdoc numbers can lead to misleading population trends
The postdoc number varies between 30,000 and 80,000 based on the survey used. One of these surveys, the Survey of Graduate Students and Postdoctorates in Science and Engineering (GSS), is currently regarded as the most comprehensive survey for assessing trends in the postdoc population. However, inconsistent reporting of institutional postdoc populations is a significant source of error in the GSS. When we analyzed the yearly changes in the biological sciences postdoc population at institutions surveyed by the GSS, we found that several institutions had widely fluctuating numbers of biological sciences postdocs, which sometimes changed by two-fold or more over a single year. Moreover, the majority of institutions were unable to explain these discrepancies for various reasons. Ultimately, the issue is that these dramatic changes mask contractions or expansions of the national biological sciences postdoc population, and therefore data reported to the GSS on this population is unreliable for assessing national trends. Thus, we propose the adoption of a unified definition of a postdoc, consolidation of postdoc titles and the creation of an index to better assess biological sciences postdoc trends without these large fluctuations.
Multiple titles impede accurate counting of postdocs
Given that there are currently 37 postdoc titles (Figure 2), this makes it very difficult to determine who is actually classified as a postdoc in institutions. Recommendations for institutions to follow in consolidating postdoc titles were recently made, with the idea that this would lead to improved training, compensation and benefits for postdocs. Two institutions were presented as examples for having been able to address these challenges, which can provide a positive model for other institutions to follow, in order to ultimately make the scientific enterprise more sustainable.
Monitoring the compliance of the academic enterprise with the FLSA
The Fair Labor Standards Act (FLSA) establishes minimum wage and overtime pay for employees. In 2016, we monitored the compliance of the academic enterprise with the FLSA as related to postdoctoral salaries. On December 1, 2016, the FLSA was due to be updated by the U.S. Department of Labor, with key changes that included increasing the salary threshold for exemption from overtime for working more than 40 hours per week, and indexing the salary level so that it is updated automatically every 3 years. These changes were due to significantly affect the postdoc population. On November 22, 2016, an injunction was granted nationwide, delaying implementation of the updates, which were finally struck down entirely on August 31, 2017. We checked university websites or contacted the HR departments of 340 institutions from the 2014 NSF Survey of Graduate Students and Postdoctorates in Science and Engineering, both before and after the injunction, to find out whether institutions planned to raise postdoc salaries. While this was the case for a number of institutions following the FLSA, it also resulted in some cancellations in postdoc salary raises following the injunction (Figure 3). To document these findings, we also created an online source showing institutional compliance with the FLSA on the Future of Research website. In addition, we obtained quotes from postdocs affected by the FLSA and injunction either in response to the salary raise and salary cancellation in blog posts.
Obtaining postdoctoral salaries using Freedom of Information requests
Following our FLSA studies, we wanted to see if we could obtain actual postdoc salary numbers from institutions. We submitted Freedom of Information Requests to all U.S. public institutions estimated to have at least 300 postdocs according to the National Science Foundation’s Survey of Graduate Students and Postdocs. The Freedom of Information Act (FOIA) is a federal freedom of information law that allows for the full or partial disclosure of previously unreleased information and documents controlled by the U.S. government. Using this mechanism, we requested annual postdoc salaries and job titles as of December 1, 2016. In this publication, we show evidence of gender-related salary discrepancies, a significant influence of job title description on postdoc salary (Figure 4), and a complex relationship between salaries and the level of total 2017 institutional NIH funding. The data obtained from these requests is also in an online resource on the Future of Research website, in order to encourage others to perform their own analyses. Overall, these results provide insights into the ability of institutions to track and report data on the postdoctoral population.
Finally, we advocate for individual reporting of postdoc salaries, and encourage postdocs to submit this information at http://postdocsalaries.com/, along with whether their salary was negotiated, which is a topic of interest also currently with very little available data. In the future, we hope that analyzing these data will provide potential ways to learn more about the postdoc population from self-reported salaries.
Conclusions
Still very little data exist on postdocs, and the conundrums related to this population are the same as they were 50 years ago. We still don’t know with certainty how many U.S. postdocs there are, and institutions are unable to explain postdoc numbers (as evidenced by those reported to the GSS in our study). Currently, also, multiple titles under which postdocs are employed impede their accurate counting and classification, meanwhile also reflecting postdoc salary disparities based on these titles (Figure 5). Overall, these data indicate the ability of institutions to collect and report postdoc data, which thereby influences the scientific enterprise as whole.
This post is adapted from a talk given by Dr. Adriana Bankston to the Molecular Biology Department at the University of Southern California in Los Angeles, CA, on March 23rd, 2018. This encompasses work from three Future of Research publications, one of which was done in collaboration with Rescuing Biomedical Research. Dr. Adriana Bankston is a member of the Future of Research Board of Directors.