Conclusions and Recommendations
The committee's study of early research careers in the life sciences revealed a flourishing, productive research enterprise with little unemployment but with a workforce heavily concentrated in ''training" positions, such as graduate students and postdoctoral fellows. The occupants of these positions are taking longer to obtain their PhDs; they continue their training after graduate school by assuming postdoctoral positions; their tenure in these postdoctoral positions is lengthening; and when they seek out permanent positions, they face stiff competition—hundreds of applicants for a single post. The net effect of those trends is an ever-growing accumulation of highly trained young scientists in positions that were intended to be transitional. Yet these very people are essential for the accomplishment of the research that has brought so much benefit to the nation and reputation to its life-science endeavor. The committee was faced with an inherent conflict: the system is producing more PhDs than can be absorbed into the permanent workforce, and these trainees are essential to the conduct of research in US universities.
The current situation is the product of a linked education-research system that is in disequilibrium because of features that are intrinsic and structural, that are not confined to the life sciences but have parallels elsewhere in higher education, and that are likely to continue to produce the same outcomes that we have just summarized.
The situation has been building for a long time. In this country, the training of PhDs in science and the performance of scientific research are intimately linked. It has been an article of faith—at least since the 1945 Vannevar Bush report—that both the body of scientific knowledge and the aptitude of young scientists benefit from this linkage. Accordingly, because graduate students play an important role in research projects, the level of graduate enrollments has been strongly influenced by growth in the research enterprise. The arrangement served the nation and the people involved very well during the period of rapid growth in the academic sector that began in the late 1950s. New programs, new departments, and new universities were eager to hire new PhDs (and these new units soon began graduate education programs of their own). By the middle 1970s, however, the growth in the system had begun to slow and it has never regained its earlier rate. Yet the number of new PhDs per year continued to rise (albeit at a much slower rate) while new academic jobs became scarcer. As those two trends continued through the 1970s and the early 1980s, the term of predoctoral study began to lengthen and the proportion of new PhDs who took postdoctoral appointments began to increase, as did the length of time they spent in that status—a sign of the imbalance. To be sure, a substantial increase in hiring in the pharmaceutical and biotechnology industries for a period in the 1980s helped to absorb some of the excess of trained scientists, but that too slowed by the end of the decade. The current situation has been exacerbated by a dramatic 42% increase from 1987 to 1996 in the annual number of PhDs awarded in the life sciences, a substantial proportion of which were awarded to foreign-born candidates. In the same period, the size of the postdoctoral pool grew as well, augmented by an influx of foreign-trained scientists.
Most of the stakeholders in the life-science community are well served by the present arrangements and are likely to be satisfied with how the system is working. The principal exceptions are the senior graduate students and the postdoctoral fellows who are searching for
research jobs with career-ladder prospects in academe, industry, or government where they can apply their lengthy training and experience. The search is perhaps most difficult for those who aspire to the university positions toward which their mentors and the academic culture guided them. Although the academic sector is the largest employer of life scientists, the number of openings there and the growth in new positions were being outstripped by the growth in the applicant pool.
Is there any need to intervene, to attempt to redress the imbalance in the system? Some say No—the system is Darwinian, and the competition for occupational survival will bring the fittest to the top. Indeed, the system is designed to winnow out the less competent; not everyone has the talents to become an independent investigator, and it is assumed that some fraction of the graduates will eventually decide to pursue other careers. The system is functioning as it should, and market forces should be allowed to prevail.
This committee takes a different position. We believe that the current rate of production is too high and certainly should not grow higher. The system of training and research that worked so well in times of overall expansion of the enterprise is increasingly deleterious in an era of little growth. The aging of the "young" scientist is disquieting. The system is delaying independence and muffling creativity at perhaps the most productive phase of the individual scientist's life. Finally—and most important—the committee is concerned that an unduly crowded labor market with small chances for success could in the long run drive out the most talented and ambitious aspirants, who will opt for more promising career opportunities in other fields and professions. When the system produces an imbalance like the contemporary one, it is inefficient, wasteful, and dispiriting to its recruits.
For those reasons, the committee believes that there is justification for intervention to adjust the imbalance in the education and training system. At the same time, we recognize the complexity of the system and the diffuse interdependence of its components. In the sections that follow, we report a variety of strategies that the committee has considered for making adjustments, asking of each strategy not only what good purposes it might serve but also what ramifications, especially unwanted consequences, it might have. We have grouped the strategies according to what we believe are desirable goals for making a start on alleviating current difficulties. Overall, our aim is to ensure the continued health of the research enterprise while confronting the disequilibrium that has created a crisis of expectations in the young cohorts who represent the future of life science. We hope that our analysis will focus on the systemic factors that led to the present dilemma and will stimulate widespread discussion in the scientific community about desirable changes.
Restraint of the Rate of Growth of the Number of Graduate Students in the Life Sciences
Over the last 2 decades, there has been a substantial growth in the number of life scientists in all categories of impermanent employment1 owing in no small measure to a sharply increasing number of PhDs being awarded by US universities to both US citizens and foreign nationals, especially in the last decade. This
growth, which has outstripped the small increases in the number of permanent positions available, has been a major contributor to the swelling of the postdoctoral pool of life scientists. The pool numbers about 20,000, many of whom are marking time until they can move into permanent positions.
Recommendation 1: The committee recommends that the life-science community constrain the rate of growth in the number of graduate students, that is, that there be no further expansion in the size of existing graduate-education programs in the life sciences and no development of new programs, except under rare and special circumstances, such as a program to serve an emerging field or to encourage the education of members of underrepresented minority groups.
The current annual rate of increase in awards of life-science PhDs—5.1% from 1995 to 1996—if allowed to continue, would result in a doubling of the number of such PhDs in just 14 years. Our analysis suggests that that would be deleterious to individuals and the research enterprise. The committee recognizes that the number of PhDs awarded each year might already be too high. Although a return to pre-1988 levels of training might be beneficial, we believe that a concentrated effort to reduce the size of graduate-student populations rapidly would be disruptive to the highly successful research enterprise. The professional structure of life-science research requires the services of graduate students and postdoctoral fellows to conduct the research that is now being funded. A serious reduction in this labor force would impair, delay, or forestall the accomplishment of current and future research.
We caution that it will be necessary to distinguish among fields when making decisions about optimal numbers of graduate students. As shown in chapter 2, almost all the increase in life-science PhD production has been in biomedical fields. Actions taken in one field of the life sciences might be unnecessary in others. It is worth noting, however, that the data shown in figure 3.10 suggest that biomedical and nonbiomedical life-science fields are experiencing similar changes in employment trends, for example, smaller fractions of PhDs finding permanent employment in academe.
The committee acknowledges that its recommendation to constrain further growth will not be easy to implement. Life-science faculties need teaching assistants and research assistants, and limiting the number of entering graduate students will be resisted. But the current rate of growth can no longer be justified, and the premises that have produced it must be reexamined. The committee urges life-science faculties to seek alternatives to these workforce needs (see below in this chapter).
The committee examined several approaches to stabilizing the total number of PhDs produced by life-science departments beyond the first and obvious approach of individual action on the part of graduate programs to constrain growth in the number of graduate students enrolled. As the increases over the last decade, as shown in chapter 2, have been fueled primarily by the increased availability of federal support for research assistants, federal agencies might restrict the numbers of graduate students that they support through the research grant mechanism. If further restrictions were placed by the National Institutes of Health (NIH) on the total amount of salary and tuition support provided for students on research grants well below the current $23,000 cap, it could reduce the attractiveness of research grants as a means of supporting graduate students, although it might also penalize many outstanding programs in private institutions that have high tuitions. Before any action of this sort is adopted, the federal agencies must carefully consider what impact it is likely to have on the university departments and the research efforts being supported.
An alternative approach to restraining the rate of PhD growth is to try to influence career decisions made by prospective graduate students. That could be accomplished, at least in part, by providing accurate and up-to-date information about job prospects for those considering careers in the life sciences. To be sure, the career choices made by students are individual decisions based on a variety of factors, including the attractiveness of alternative career opportunities, the availability of financial support, and a host of personal circumstances. Nevertheless, the most prudent way to reasonably reduce the rate of increase in the number of PhDs awarded annually and perhaps to achieve a gradual reduction in the numbers being trained is to help students to make informed decisions about their career choices. The kinds of information that might be provided and how it might best be compiled are discussed in the next section.
Dissemination of Accurate Information on the Career Prospects of Young Life Scientists
Recommendation 2: The committee recommends that accurate and up-to-date information on career prospects in the life sciences and career outcome information about individual training programs be made widely available to students and faculty. Every life science department receiving federal funding for research or training should be required to provide to its prospective graduate students specific information regarding all predoctoral students enrolled in the graduate program during the preceding 10 years.
Several groups have recognized the need to provide prospective graduate students accurate and up-to-date information on career prospects. As early as 1982, a National Research Council committee studying the employment opportunities for postdoctoral fellows in all fields of science and engineering recommended that the National Science Foundation (NSF) expand its national data-gathering effort to include a survey specifically focused on career decisions of young scientists and engineers. In 1995, a report of the National Academy of Sciences' Committee on Science, Engineering, and Public Policy on graduate education in science and engineering concluded that academic departments should provide employment information and career advice to prospective and current students in a timely manner. Despite those and many other calls for better career information, most life-science students today must rely primarily on the anecdotal reports of their mentors and fellow students.
The earlier recommendations stressed the importance of information for current and prospective graduate students but this committee believes that such data would be equally valuable to faculty, university administrators, and federal policy-makers. In particular, the committee is concerned that the goals discussed here might never be achieved unless the entire life-science community understands fully the implications of the employment trends.
The committee has considered several options to achieve the goal of improved career information. The first is to disseminate widely the data presented in this report. Chapter 3 and the appendixes contain a wealth of information about employment trends over the last 2 decades for young PhDs in the life sciences. Nevertheless, these data have important limitations. First and foremost, because the findings from the Survey of Doctorate Recipients are based on less than 10% of the PhD population, reliable estimates are not available for graduates in a particular discipline, department, or ethnic group.
Thus, although the demonstrated global trends could be useful to policy-makers, they are not especially helpful to faculty advisers and their students who are considering individual career decisions.
A second option would be to expand the sample of recent graduates included in NSF's national survey. Because in recent years this survey has obtained a relatively high response rate (greater than 80%), an expansion of the sample might be expected to yield high returns. The committee regards this step to be valuable but it might not be sufficient to meet all the information needs. For example, reliable data on the early careers of graduates from particular departments would not be available unless a very large sample of recent graduates were selected—and the costs of such a large sample would probably be prohibitive.
A third option that the committee strongly endorses would be to require every department that receives federal funding for research or training to provide current employment information on all predoctoral students enrolled in its program during the preceding 10 years. Such information might include
- The number of trainees and their sex, citizenship, and ethnicity.
- The number of students who left the program before completing their training.
- The length of time from enrollment to degree for each student.
- The current employment situation of each graduate.
One of the major obstacles in implementing a national data collection of such magnitude would be making certain that all federally supported departments provide accurate and comprehensive information that is in a standard format so that comparisons among different departments can be made. Although the difficulty of obtaining reliable information on the current employment situations of graduates from 10 years earlier should not be underestimated, the task is feasible, as demonstrated by the fact that this information has long been a standard requirement for university programs applying for NIH training grants.
A fourth option would be to ask professional societies to assume greater responsibility for compiling and disseminating early-career information. In several science fields (such as chemistry, mathematics, and physics), the professional society conducts a survey of recent doctorate recipients and reports median starting salaries, unemployment rates, and other market indicators. Such a survey would be more difficult in the life sciences because no professional society covers all the disciplines. Nevertheless, professional societies in the life sciences could play active roles in disseminating the information collected by any of the approaches described above. And indeed the committee notes that the Federation of American Societies for Experimental Biology has already published some findings from an analysis similar to that presented in chapter 3 of this report.
Improvement of the Educational Experience of Graduate Students in the Life Sciences
In addition to its interest in constraining the further growth of PhD output, the committee was concerned about aspects of the current system of supporting graduate training, especially the growth in the fraction of graduate students who are employed as research assistants by the research grants of their mentors. The federal government supports about one-third of all life-science graduate students at any time and about
two-thirds at some time in their training, most through salary and tuition provided in the research grants of faculty mentors. That category of student support accounted for the largest percentage of the increase in graduate-student enrollment over the last decade.
There is no clear evidence that career outcomes of persons supported by training grants are superior to those of persons supported by research grants (see discussion in chapter 5). However, the committee, which included members with direct experience with training grants, concluded that training grants are pedagogically superior to research grants and result in a superior educational climate in which students have greater autonomy. First, training grants are pedagogically superior because they provide a mechanism for stringent peer review of the training process itself, something that is not considered in the review of a research project. Second, they improve the educational climate because they minimize the potential conflicts of interest that can arise between trainers and trainees. Although the student-mentor relationship is ordinarily healthy and productive for both partners, it can be distorted by the conditions of the mentor's employment of the student and limit the ability of students to take advantage of opportunities to broaden their education. Third, training grants provide the federal government with information that it needs to evaluate the level of its investment in graduate life-science education with the aim of developing a funding framework for graduate education that contributes to the long-term stability and well-being of the research enterprise.
Recommendation 3: The committee encourages all federal agencies that support life-science education and research to invest in training grants and individual graduate fellowships as preferable to research grants to support PhD education. Agencies that lack such programs should look for ways to start them, and agencies that already have them should seek ways to sustain and in some instances expand them.
This recommendation should not be pursued at the expense of scientific and geographic diversity. Rather, we encourage the establishment of small, focused training-grant programs for universities that have groups of highly productive faculty in important specialized fields, but might not have the number of faculty needed for more traditional, broad-based training grants.
It is true that the current regulations governing NIH training grants bring universities some financial disadvantages because of restricted overhead recovery. Furthermore, NIH training grants cannot support foreigners on student visas, and so this recommendation places at disadvantage programs that depend on foreign students for research or teaching. These disadvantages are outweighed, in the committee's view, by the salutary effect that the training-grant peer-review process brings to the members of a department faculty, leading them to examine and reflect on how, as an entity, they are providing for the education and training of their graduate students.
Our endorsement of training grants and fellowship is not intended to result in the training of more PhDs, which we argue would be entirely inappropriate. Rather, any growth in the numbers of trainees supported through an expansion of training grants should come at the expense of the numbers of trainees supported on research grants. Thus, the implementation of this recommendation should produce no increase in the numbers of students but only a change in the mechanism by which their training is supported by federal funds. It would be best if principal investigators voluntarily reduced the number of students they support on their research grants as support via
training grants grew. However, NIH, the largest provider of both training grants and research grants, and other agencies would be required to manage the numbers supported by research grants to achieve the committee's goal of constraining further growth.
The committee is also concerned that the length of time spent in training has become too long, at a median of 8 years of elapsed time from first enrollment to PhD in all the life sciences though field differences exist. We believe that the time should be about 5–6 years. However, an immediate effort to shorten the time to degree would increase the number of PhDs produced. Efforts to shorten the time to degree should be undertaken when the effort to restrain growth in the number of PhDs has shown positive effects.
Enhancement of Opportunities for Independence of Postdoctoral Fellows
While the length of graduate training has been increasing, so too have the extent and duration of postdoctoral training. Prolonged tenure as a postdoctoral fellow provides a person with valuable research experience, but it carries some real costs. In most cases, fellows are not independent of their mentors so they can not pursue their own research. We recognize the many good reasons for prolonged tenure as a postdoctoral fellow but we believe that tenures longer than 5 years are not in the best interest of either the individual fellow or the scientific enterprise.
Unfortunately the committee did not identify a way to rapidly achieve a reduction in the tenure of postdoctoral fellows. The lengthening of the postdoctoral period seems to be due largely to the highly competitive job market for permanent positions in academe and industry; the situation will change only if there is an increase in the number of new positions or a decrease in the candidates for them.
Recommendation 4: Because of its concern for optimizing the creativity of young scientists and broadening the variety of scientific problems under study in the life sciences the committee recommends that public and private funding agencies establish "career-transition" grants for senior postdoctoral fellows. The intent is to identify the highest-quality scientists while they are still postdoctoral fellows and give them financial independence to begin new scientific projects of their own design in anticipation of their obtaining fully independent positions.
The recommendation is based on the experience of the Lucille P. Markey Charitable Trust's Scholars in Biomedical Sciences Program, which until recently supported 16 postdoctoral fellows per year for 2 years of additional postdoctoral work and 5 years as faculty members. Although the program was very small, it identified excellent candidates relatively early in their careers and gave them financial and intellectual independence. Not surprisingly, the Markey scholars were very successful in obtaining permanent tenure-track positions in academe. Since the termination of the Markey program, the Burroughs Welcome Fund has established a comparable program for life scientists. A program administered by the US Department of Agriculture provides postdoctoral fellows the opportunity to apply for research grants and perform independent research.
We propose grants of 4–5 years in duration that would provide senior postdoctoral fellows (those with more than 2 years of postdoctoral experience) salary commensurate with their experience and a modest supply budget. Successful proposals would define an innovative research project that was distinct from the work going on in the current mentor's laboratory. A mentor would provide laboratory space and would
acknowledge in the applications that the project was the intellectual property of the applicant and would leave the laboratory when the applicant did.
The committee recommends a goal of 200 federal grants awarded annually, representing about 1% of the postdoctoral pool. That number of people supported would be quite small but the program might provide an important opportunity for the most promising postdoctoral fellows and serve as both example and incentive to many more. We make this recommendation with the knowledge that it is possible that the money for a new federal grant program probably would come from existing federal funds. In our view, the benefits of increased intellectual independence and improved motivation of talented midcareer post-doctoral fellows justify such a reallocation of funds. Private funders might establish new programs or enlarge existing programs that support career-transition grants.
The career transition grant would differ from existing federal research grants in several important ways. First, permission to apply for traditional grants is usually restricted by institutions to principal investigators who have some form of faculty status, whereas these new grants would go to postdoctoral fellows. Second, the career-transition grants would be modest in scale and would not provide salary support for other laboratory personnel or trainees. Finally, the grants would be transferable to new host institutions once the applicants obtained positions and would terminate on receipt of faculty awards. The success of this recommendation depends on a willingness of training institutions to accept grants to persons who do not have faculty status at the time of application.
The benefit of career-transition grants to individual young scientists is obvious: increased independence means increased opportunity to pursue novel ideas and to make progress in work that can establish a career, opening opportunities for future independent employment. Substantial benefits would also be realized by the scientific enterprise as a result of this stimulation of research energy and the increased diversity in the scientific ideas being pursued. Less obvious but no less important is the benefit that would accrue to the mentors. The presence of more experienced scientists in the host laboratories, although not directly contributing to the productivity of the mentors' work, will contribute to the intellectual climate of the laboratories.
Alternative Paths to Careers in the Life Sciences
As traditional research positions in academe, industry, and government have become more difficult to obtain, positions in "alternative careers"—such as law, finance, journalism, teaching, and public policy—have been suggested as opportunities for PhDs in the life sciences.
The idea of highly trained PhD scientists investing their talents in nontraditional careers seems at first glance attractive. Scientists have analytic skills and a work ethic to bring to any position, and the placement of highly trained scientists in diverse jobs in the workforce would lead to an increase in general science literacy. As the committee's review of alternative opportunities (chapter 4) concludes, however, most of the possibilities are less available or less attractive than they might at first glance appear. Many "alternative" careers are also heavily populated, and competition for good positions is stiff. Others require special preparation or certification or offer unattractive compensation, and none makes full use of the PhD's hard-won life-science research skills. The committee believes that the idea of alternative careers should not be oversold to PhD candidates.
The interest in alternative careers for PhD scientists has inevitably raised the question of whether preparation for the degree should be changed from its current narrow focus on training for the conduct of scientific research to embrace a broader variety of educational goals that would connect to alternative career paths. The committee has discussed that question extensively.
Recommendation 5: The committee recommends that the PhD degree remain a research-intensive degree, with the current primary purpose of training future independent scientists.
We have several reasons for that recommendation. First, a steady supply of new, highly trained investigative talent is essential for maintaining the growth and vigor of life-science research and for exploiting the opportunities of future discoveries. Second, the majority of people so trained are using their skills and abilities in life-science positions. Third, we have not been able to identify a substantial number of unfilled opportunities in alternative careers.
At the same time, the committee recognizes that not all students who begin graduate school intending to pursue research careers maintain that desire as they progress through training. Graduate programs should expand their efforts to help students to learn about the diversity of career opportunities open to them, and university departments should examine possible alternatives to the research PhD, for example, rigorous master's-degree programs in applied fields of the life sciences.
The master's degree might be a more appropriate and point for students who determine early enough in their training that PhD training is not necessary for the career goals that they have selected. There has been a decline in the number of master's-degree programs in the life sciences and with it a growing perception that the master's degree has become a consolation prize for those who do not complete a PhD program. Those changes effectively limit the number of choices for college graduates who are interested in a career in the life sciences, although not necessarily careers in directing laboratories conducting fundamental research.
Recommendation 6: The committee recommends that universities work to identify specific fields of the life sciences for which master's-degree training is more appropriate, more efficient, and less expensive than PhD training and that focused master's-degree programs be established in those fields.
A reinvigoration of the master's degree will require that new programs be intimately tied to the opportunities in the labor market. For example, a life scientist who is interested in a K-12 or 2-year-college teaching career would benefit from formal and focused master's-degree programs that do not require long periods of research-intensive graduate and postdoctoral training. In chapter 4, we report that life-science PhDs have not been prone to take positions as precollege teachers. Certainly, there is a need for persons with life-science knowledge to enter teaching careers. Intensive efforts are under way to change the nature and extent of science education in our schools. Those efforts, based on the National Science Education Standards and similar reform documents, emphasize teaching science as inquiry rather than as word associations. None of this will be possible without a structural change in the profession of precollege teaching and a large cadre of people who both understand science and the nature of science as inquiry and have been trained as lead teachers and science-resource specialists. Focused and intensive master's-degree programs would be not only more appropriate but also preferable to the PhD for this type of employment.
Interdisciplinary master's-degree programs
might combine advanced life-science training with studies in nonscientific fields—such as management, public affairs, and engineering—that would prepare candidates for positions in government and industry. A vigorous master's-degree program that produces highly skilled laboratory technicians for industry, government, and academe could potentially contribute to righting the imbalance between PhD training and the labor market. When the committee recommended constraint in further growth in training in recommendation 1, it was fully aware that graduate students are needed in the labor-intensive life-science research enterprise and to teach undergraduates. One way to resolve this dilemma is to effect a modest shift toward a more permanent laboratory workforce by replacing some fraction of the existing training positions with permanent employees, such as MSc-level technicians and PhD-level research associates.
A system of that kind, with less reliance on trainees to conduct research, has been in operation in Europe for many years. Nevertheless, there is likely to be strong resistance to such a change in the US scientific community. Permanent employees would require better compensation in the form of salary and benefits than graduate students and postdoctoral fellows and could not be expected to work the long hours of most trainees. As a consequence, a shift to a more permanent workforce would probably result in some reduction in productivity and cost effectiveness. Furthermore many US scientists are of the opinion that the creativity of US science comes from the young and inquiring minds of young trainees. Despite those reservations, the committee believes that a broader discussion of this option within the life-sciences community is warranted.
The Impact of Foreign Nationals
This report has documented that much of the recent increase in the number of life-science PhDs granted by US universities are foreign nationals, not US citizens—in some years, as much as one-fourth of the degrees awarded. The number of foreign nationals reflects the international nature of modern science and the central place that the United States plays in this international arena. Furthermore, foreign nationals have traditionally contributed to the excellence of US science, as suggested by the fact that of the 732 members of the National Academy of Sciences who are life scientists, 21.2% are foreign-born and 12.4% obtained their PhD training abroad. Foreign nationals' important contributions to US scientific leadership is reflected in their inclusion as "outstanding authors" in life sciences (26.4%). Foreign students and fellows are welcome participants in the research enterprise, provided that they are of high quality and competitive with American applicants.
Although the reasons for the increase in degrees awarded to foreign nationals are not altogether clear, the committee understands that it is a phenomenon essentially controlled by life-science departments themselves, inasmuch as immigration law virtually delegates visa decisions to universities. Departments and universities make their own admission and funding decisions and issue documents to those they admit, which nearly always results in the US government's issuing student visas (subject to checks for fraud and funding availability). The freedom given to US universities to determine how many foreign students they will admit carries responsibilities. If misused, it could vitiate the committee's recommendation to provide up-to-date and full career information to prospective applicants for graduate education in the life sciences. That information might have a powerful effect on US citizens but it is highly unlikely to have a similar effect on students from low-wage economies with poor educational or research opportunities. Even the low stipends paid to graduate students enable a higher standard of living for such applicants; and the prospect of a job or postdoctoral position and a permanent visa at the completion of
graduate study is a powerful incentive for citizens of many countries.
We believe it would be unwise to place arbitrary limits on the number of visas issued for foreign students. But we do not believe that US institutions should continue to enroll unlimited numbers of foreign nationals. As decisions are made on ways to constrain further growth, the measures adopted should apply equally to all students regardless of nationality.
Recommendation 7: If, as we hope, implementation of our recommendations results in constraining further growth in PhDs awarded in the life sciences, we urge our colleagues on graduate admissions committees to resist the temptation to respond by simply increasing the number of foreign applicants admitted.
Postdoctoral fellows are also recruited from abroad. At present, half the roughly 20,000 postdoctoral fellows in the United States are foreign nationals, many of whom entered the country with PhDs awarded elsewhere. These scientists constitute an important part of the research labor force, as well as of the pool of applicants for permanent jobs in academe, industry, and government. In this instance again, we urge our colleagues to give equal opportunity to US citizens and foreigners and to refrain from hiring foreign nationals to fill the places of US scientists.
Responsibility for Effecting Change
This report has documented several dramatic changes in career trends in the life sciences over the last several decades. The rapid growth in the academic scientific establishment in the 1960s and the early 1970s set in place a training infrastructure that was built on the premise that there would be continued growth. When the inevitable slowdown in resources to support that growth occurred, it was not accompanied by a commensurate adjustment in the rate of training. The impact of the imbalance between the number of aspirants and the research opportunities is now being felt by a generation of scientists trained in the last 10 years who are finding it increasingly difficult to find permanent positions in which their hard-accumulated skills in research can be used. Unless steps are taken to put the system more in balance, the difference between students' expectations and the reality of the employment market will only widen and the workforce will become more disaffected. Such an occurrence would damage the life-science research enterprise and all the participants in it.
The training of life scientists is a highly decentralized activity. Notwithstanding the heavy dependence on federal funds, the most important decisions affecting the rate of production of life scientists are made locally by the universities and their faculties. The numbers and qualifications of students admitted to graduate study, the allocation of institutional funds for their tuition and stipends (which account for half or more of the total expenditures for graduate-student support), the requirements for the degree—all are local decisions. As a consequence, a large portion of the responsibility for implementing our recommendations falls on the shoulders of established investigators, their departments and universities, professional scientific organizations, and students themselves. Students must take the responsibility of making informed decisions about graduate study, but they must be provided accurate career information on which to base their decisions. Individual faculty members must be willing to set aside their short-term self-interest in maintaining the high level of staffing of their laboratories for the sake of the long-term stability and well-being of the scientific workforce. Directors of graduate programs must be willing to examine the future workforce needs of the scientific fields in which they train, not just the current needs of their
individual departments for research and teaching assistants.
The recommendations in this report are offered as first steps to improve the overall quality of training and career prospects of future life scientists. We hope that the information in this report will be used to begin discussions within the life-science community on the best ways to prepare future scientists for exciting careers in the profession and to protect the vitality of the life-science research enterprise.