The old adage goes, "If it ain't broke, don't fix it." But, when a system is broken, as has been shown for postdocs in multiple independent studies, how should one go about fixing the "it"?

This was the subject of a recent meeting convened by the Network on the Scientific Workforce of the National Bureau for Economic Research and the Labor and Workforce Program of Harvard Law School. In the meeting, held at Harvard University 1 to 2 November, organizers called together graduate students, postdocs, economics and biosciences faculty members, administrators, and funding-agency officials to discuss the various problems facing grad students and postdocs and the different strategies employed by these groups to find redress.

In the opening session, Harvard economist and conference organizer Richard Freeman put the postdoc compensation issue into perspective. Postdocs earning the 2002 National Institutes of Health National Research Service Award minimum of $31,092 and working 50 hours per week (or 2500 hours per year) earn $12.44 per hour, just over the starting salary of a Harvard janitor. This similarity of wage structure was also noted at Brookhaven National Labs earlier this year. Freeman wryly noted that the Harvard janitorial staff is currently campaigning for a living wage.

Looking at the demographics of the postdoc population--gleaned from the National Academies' 2000 Committee on Science, Engineering, and Public Policy (COSEPUP) Guide for Enhancing the Postdoctoral Experience, and the National Science Foundation's (NSF's) Survey of Graduate Students and Postdoctorates in Science and Engineering (the latter accessed using the WebCASPAR database)--one finds that the average age of postdocs is creeping up to 35, 60% of postdocs are married, and 40% have children. Given the lengthening time-to-degree for graduate students and the increasing time spent in postdoc positions, for many, stipends might constitute 25% to 30% of a young scientist's lifetime income! This confluence of conditions makes it clear why stipend, health benefits, and leave policies are critical to postdocs.

Why is it that postdocs are only recently beginning to organize themselves to bring these issues to light? In large part, the move to organize has been fueled by decreasing academic job opportunities. A 1998 American Society for Cell Biology (ASCB) survey showed a marked change in the perception of post-postdoc job prospects. Ninety percent of those currently overseeing graduate students and postdocs think it is more difficult to obtain an independent position in biology than when they were first starting out. Seventy-nine percent agreed that too many Ph.D.s were being trained for too few independent positions. And 61% of the respondents thought that funding is now much more difficult to obtain.

Certainly, proportionately fewer NIH grants are going to young researchers (see The Aging of Young Scientists). In 1980, 23% of grants went to researchers aged 35 and under; in 2002, that number had decreased to 4%. The hurdles facing young scientists seeking grant funding are growing increasingly higher, as, it appears, are the hurdles in the tenure process. Overall, numbers from NSF's Survey of Earned Doctorates, which follows a cohort of Ph.D.s for 10 years, show that in 1973, 74% of natural science and social science doctorates were in faculty positions 1 to 3 years post-Ph.D. In 1999, those numbers had decreased to 37%. Between 1987 and 1999, there has been a large increase in the ratio of postdocs to faculty positions as well--a change primarily affecting the life sciences. (Postdoc counts are from NSF's Survey of Graduate Students and Postdoctorates in Science and Engineering; faculty numbers are from SESTAT.)

Yale graduate student and Graduate Employees and Students Organization representative Maris Zivarts presented data on this phenomenon during the meeting (see Tables I-1 and G-4 in the Yale Book of Numbers). The number of senior faculty positions in the Arts and Sciences college at Yale has remained fairly constant since 1992, the number of junior faculty members has decreased by over 20%, and the number of postdocs has increased by over 200%.

Freeman emphasized that unless there is a huge upswing in hiring by universities, there will be no room in academia for the vast majority of natural sciences Ph.D.s who desire faculty positions. What about jobs outside of academia? According to the National Academies' COSEPUP Guide for Postdocs, 2000, the number of postdocs employed by industry and government has increased about fivefold between 1989 and 1997. However, the market for post-postdocs outside of academia is uncertain.

So, given that there is a growing number of postdocs, how is it that they have been able to develop any momentum to change their working conditions? The answer is: Because there is no oversupply of postdocs, just an undersupply of post-postdoc positions. NIH reports a 5% per year increase in demand for postdocs. Many faculty members are struggling to staff their labs and are increasingly turning to international scientists--which might explain why less than 40% of U.S. graduate students are non-U.S. citizens, but over 50% of postdocs are from abroad (NSF Survey of Earned Doctorates, 2001).

Thus, at many institutions, postdocs find themselves in a strong position to negotiate for improved working conditions. Some institutions, such as the University of Alabama, Birmingham, and the University of Pennsylvania, have begun to initiate reforms even before postdocs formed associations. In the same time period, we have seen the implementation of an increased postdoc stipend scale at NIH and movement within NSF to begin an examination of postdoc policies.

Over the past 3 years, the number of postdoc organizations has increased to more than 80 worldwide. In the United States, at last count, 74 institutions housed a postdoc association and/or office. At the meeting, Karen Christopherson, a Stanford postdoc, and James Nelson, a Stanford professor, explained how, at their institution, coordinated efforts between postdocs, faculty, and administrators have been effective in improving the working conditions for postdocs, including a minimum salary, a grievance policy, and term limits for postdocs.

Raymond Clark, a postdoc at the University of California (UC), San Diego, described how postdocs at the nine UC campuses have created a systemwide postdoc council. The UC postdocs created the council because they found that some policies couldn't be addressed at the local campus level. This council has been working with the UC system administration to develop policies for postdocs. It has negotiated minimum salary guidelines, a policy the UC central administration appears committed to enforcing. At a recent council meeting, postdoc representatives negotiated to include vision and dental insurance in the new UC-wide postdoc health policy. At the very least, the council provides a clear venue for postdocs and the UC system to interact on issues of common concern.

Following the 2002 Postdoc Network meeting, at which postdocs and others discussed their common concerns, Orfeu Buxton, a postdoc at the University of Chicago, told meeting participants that further progress required the creation of a national postdoc association. As the UC postdocs realized, some policies can be addressed only at the national level. Buxton joined with postdoc leaders from several institutions to draft a business plan and funding proposal that is currently under review. They hope to establish a self-sustaining organization that will focus on the national policy interests of postdoctoral scientists.

Among the primary goals of this group are gathering information that may inform a consensus on "best practice" policies, developing educational initiatives to disseminate these policies to postdoctoral scientists and institutions, and encouraging and facilitating their implementation. The nascent organization also intends to enter into meaningful dialogue with the federal funding agencies and professional organizations to advocate for policies at the national level, and it will actively track progress from the local to federal levels.

The question remains, though, what about post-postdoc opportunities? Should postdocs continue to be trained for academic jobs that for the most part do not exist?

The current training system needs an overhaul, according to Frank Solomon, a professor of biology at the Massachusetts Institute of Technology. Solomon cited the increase in the number of people listed as authors on research papers as evidence that the apprenticeship training model is no longer applicable to postgraduate research in the United States. The one-on-one relationship between postdoc or grad student and faculty mentor is becoming rare. It would seem that the relationship between postdoc and faculty is shifting from trainee to employee.

At the end of the day, with the number of postdocs outpacing growth in the academic sector, many of the conference participants called for a change in the training paradigm--broadening accepted post-postdoc employment possibilities. This feeling corroborates findings from the ASCB survey, where an overriding majority of respondents called for institutions to offer a mixed curricula to prepare Ph.D.s for a variety of research and nonresearch careers.

Beyond institutions, redirecting training efforts will require a concerted effort on the part of postdocs, faculty, disciplinary societies, and funding agencies to provide information on the variety of jobs that trained scientists are qualified for and professional development programs to help Ph.D.s gain entry into these fields.