Trends in the Early Careers of Life Scientists (1998)

The future promises many exciting opportunities for scientific research in the life sciences, but there are also considerable uncertainties. This chapter briefly identifies some of the newly emerging fields of the life sciences that hold particular promise for the immediate future. It then describes some of the uncertainties that life scientists will face and concludes with a discussions of the diversity of career options that might be available to young life scientists now and in the future.

Research in the life sciences is high on our nation's list of priorities largely because of the likelihood that this research will improve the well-being of our population. Of the many promising fields of science that will contribute to economic and social well-being, we mention here only a few examples.

The 1990s have been called the "decade of the brain", and neuroscience offers essentially unlimited challenges and opportunities in both basic and applied research. High on the list of promising fields of research is the quest for links between cognition and the molecular activity of memory processes in the brain. New concepts and new techniques are opening exciting research opportunities. For example, neuroscientists are using state-of-the-art genetic engineering, imaging methods, and monitoring of brain-cell physiology to define the molecular bases of memory, recognition, and learning in experimental animals. The molecular mapping and elucidation of complex brain-cell functions will advance the understanding of Alzheimer's disease, learning disorders, addiction, and other medical and psychological conundrums that currently plague society. Careers in the neurosciences can be based on training in many combinations of molecular biology, neurobiology, physiology, psychology, and computer science.

Gene therapy is based on the transfer of genetic material into a human. Gene delivery can be accomplished either directly by the administration of gene-containing viruses or DNA to blood or tissues or indirectly through the introduction of cells that have been manipulated in the laboratory to harbor foreign DNA for the purpose of treating disease. By altering the genetic material of somatic cells, gene therapy could correct underlying disease-specific pathophysiologic characteristics. In some instances, it offers the potential of a one-time cure for devastating, inherited disorders, such as diabetes. In principle, gene therapy should be applicable to many diseases for which current therapeutic approaches are ineffective or when the prospects for effective treatment appear exceedingly low. As of June 1995, 106 clinical protocols involving gene transfer had been approved by the National Institutes of Health (NIH) Recombinant Advisory Committee (RAC). Indeed, more than 600 human subjects have already undergone gene transfer experiments. NIH provides about $200 million per year for research related to gene therapy, and industrial support of gene-therapy research has grown steadily. Industry now exceeds NIH in funding and underwrites most of the approved clinical protocols. This young field is a frontier of modern medicine, open to people with MD or PhD degree in molecular genetics, molecular biology, or related sciences.

All of genetic information in an organism is encoded in the DNA or RNA sequence of its genome. The genome projects that are now under way are producing vast amounts of data that will be essential for understanding the normal and pathologic physiology of humans and of the many plants and animals on which our lives depend. There are, however, many unsolved problems related to genome research, some of which are so novel that they are only now being defined as specific subjects for research. For example, how is gene expression regulated on the molecular level? How does chromosomal architecture influence the rate of gene expression? How is the three-dimensional structure of proteins defined by the amino acid sequences that are specified by the genome? What are the mechanisms of protein-protein recognition in complex biochemical processes? What processes regulate the assembly of protein complexes into organelles?

Structural biology provides some of the research tools that are necessary to solve those grand challenges in molecular and cellular biology. Current research is providing improved techniques by which to determine the high-resolution structures of macromolecules, and these methods are being used to study processes of molecular recognition, signal transduction, allosteric regulation, and protein folding. The resulting data are often of immediate practical value for such undertakings as rational drug design. They are also of fundamental theoretical value as thermodynamic and kinetic data become available to complement the structural information. The resulting synergy between different kinds of molecular data is providing the views that will be necessary to understand complex biologic processes. This critically important line of inquiry is now in its earliest stages, and considerable effort will be required to realize the practical benefits of such research. A person interested in a career in structural biology should obtain a PhD degree in biochemistry, biophysics, or structural and computational biology. Prerequisites include a strong background in computer science and physics, chemistry, biology, or mathematics.

Bioinformatics uses computer technology to solve informational problems in the life sciences, for example, the identification of DNA sequences in the human genome that are markedly similar to genes that have been identified and studied in experimental organisms such as yeasts. The computer databases of genome and protein sequences are now large enough to require new models for the analysis and comparison of biologic systems, and new algorithms are under development to integrate heterogeneous data into coherent programs. Informatics also plays a role in modeling the interactions between drugs and proteins or physiologic processes, in the diagnosis of disease, and in keeping track of huge databases, from the DNA sequences cited above to records of patient care.

Medicine is an information-based art and science, and the opportunities for computer applications are constantly expanding. Three-dimensional visualization of human anatomy is already an instructional tool, and the visual modeling of changes in tissue structure during disease progression offers parallel opportunities. Large pharmaceutical houses are especially interested in scientists with training in bioinformatics, given the explosion of new data from large-scale sequencing projects, like the work on the human genome, which will require new technologies for information processing to assist in the exploitation of data for product development. Young people with advanced training in statistics, information theory, artificial intelligence, and other aspects of computer science can make major contributions.

The growth of human populations is an important driving force in the accelerating changes that are occurring in the managed ecosystems on which we depend for food, fiber, and services, such as the maintenance of clean air and water. Human activities are measurably changing the composition of the atmosphere, adding carbon dioxide and methane, which alter the radiative balance of the planet, and chlorine gas, which destroys the ozone layers in the stratosphere. Humans have already destroyed vast tracts of tropical forests and agriculturally productive land. Industrial and human wastes have degraded some of the largest sources of fresh water. We are witnessing the rapid extinction of many species and the introduction of pests and infectious organisms into new environments, sometimes with calamitous results. There is an obvious need for increased attention to these problems and for research to find their solutions. Scholars who are expert in all aspects of environmental sciences will be required to understand the increasing stresses placed on the environment by the expanding human population and the concomitant growth of industry. Careers in this challenging field will require training in population biology, ecology, the social sciences, and related agriculture sciences.

The major increases in agricultural productivity that followed World War II were attributable in part to the widespread use of synthetic chemical pesticides for the control of insects, weeds, and plant pathogens. Initial successes have been followed by unexpected consequences, including injurious effects on nontarget organisms, contamination of soil and water with chemical residues, and the development of pesticide resistance, particularly among insects. In addition, the potential harmful effects of pesticides in the food chain offer considerable reason for concern.

There is a growing consensus that pest-management systems based on biologic control agents will provide a more desirable approach for resolving some of the current problems and reducing the use of synthetic pesticides. Achieving a shift to biologic control agents will, however, require the development of treatment strategies that are inexpensive, are easily applied, offer little or no hazard for nontarget organisms (including people), are equal in efficacy to or better than current pesticides, and are predictable under a range of environmental conditions. The successes in developing biologic control systems for insects have not been matched in progress toward commercial biologic control of plant pathogens or weeds. Unfortunately, the knowledge that is necessary to develop such biologic control agents will require a massive expansion of current research effort, and it will involve the complete spectrum of basic and applied life sciences.

Many of the major corporations involved in development of disease-control agents have closed research laboratories that have a primary assignment in biologic control agents. Emphasis has shifted to transgenic plants with insect-control characteristics or chemicals that turn on resistance mechanisms when applied to plants. Extensive growth in this type of research is foreseen. Some of the plant diseases that are most recalcitrant to all known control efforts are caused by soil borne pathogens. A deeper understanding of the complexities of the physical and biologic components of soil will require research on the microflora and microfauna of the leaf and root systems of plants going well beyond the bounds of our current knowledge. Furthermore, biologic control agents that are highly effective under greenhouse conditions are often ineffective or unpredictable when tested in the field and in different geographic regions. Thus, it is likely that extensive

field testing and modification will be needed to develop and market effective biologic products. This phase of development will require many more agricultural biologists than are available today.

A different opportunity for expanded employment of life scientists will be found in aquaculture. There has been a dramatic decline in the productivity of fisheries around the world, and successful expansion of aquaculture will depend on increased knowledge about the diseases of fish, the application of improved breeding and selection procedures, and the nutritional requirements of fish under the controlled conditions of aquaculture systems. This is a comparatively unexplored field of modern biology in which much remains to be done.

It is difficult to predict how research funding will fare in the future. Just 2 years ago, in a mood of concern about reduction of the federal budget deficit, it was predicted that the budgets of federal research agencies might fall by up to 20%. In President Clinton's proposed budget for FY 1999, the planned increase for NIH is 8.4 and the increases proposed for the National Science Foundation and the Department of Energy are even higher. It is important to note that research budgets were not static from the late 1980s to the present. NIH regularly increased its budget by about 5% per year. But chapters 2 and 3 show that the large increase in the number of life-science PhDs resulted in decreases in the fractions of the PhDs who obtained "permanent" positions in academe, industry, and government research. Whether the increases proposed for FY 1999 will come about and whether increased funding will change the trends that we have reported is problematic. The mood in Washington continues to favor containment of discretionary expenditures.

On the national level, the shifting of responsibility for welfare expenditures to the states and the states' preoccupation with healthcare costs, prison costs, and their own financial situations, imply that state support for research is not likely to expand. Indeed, state support for public higher education has moderated under all those trends, and public higher education has increasingly been financed by tuition income rather than tax revenue.

Nongovernment sources of support clearly are important for basic life-science research and funds from private foundations, such as the Howard Hughes Medical Institute and the American Cancer Society or American Heart Association, will probably continue at the same or slightly increased levels. But private philanthropy does not have the resources to compensate for a substantial decrease in federal funding (Ruzek and others 1996). Although industry now spends more on life-science research and development than does the federal government, industrial research is targeted mostly at problems that are expected to yield commercial payoffs in the short run. Only the government is currently willing to take the long-range view that recognizes the tremendous returns offered over the years by investments in basic research. The basic life-science research enterprise must therefore assume that major increases in its grant support are unlikely.

The nation's research universities face increasing financial pressures that are forcing changes in priorities and shifts of resources to different academic purposes. Of special interest

for this report is the impact of such reorganization upon university-based research in the life sciences. For the last 10–15 years, university operating costs have been rising rapidly—more rapidly in most instances than inflation (Clotfelter 1996). Every cost, from janitorial supplies to faculty salaries, has increased while increases in income have not kept pace. Below are some specific examples.

Like all institutions of higher learning, research universities have accepted the responsibility of providing financial aid to undergraduate students from minority and disadvantaged populations. Many private universities have maintained policies of need-blind admission and need-based financial aid by drastically increasing the fraction of their resources that is devoted to this purpose. Except for the few universities that have very large per-student endowments, the funds for financial aid have come mainly from increases in tuition. Reliable studies estimate that 15–40% of tuition revenue is used for undergraduate financial aid at various private institutions. The steep increase in tuition has, however, begun to arouse public concern, if not resistance, and has put pressure on universities to limit future increases. Tuition at public universities too have been rising faster than inflation, as the share of educational costs supported by state governments has declined.

Increased attention to undergraduate education at research universities has resulted not only from these financial factors, but also from evidence that their clientele is becoming aware that some portion of undergraduate tuition has implicitly subsidized research. The intellectual justification for this subsidy is that undergraduate access to leading researchers is a unique feature of research universities. It follows that providing an attractive environment for research-oriented professors is a legitimate part of the cost of undergraduate education. The question remains open whether families will continue to accept this rational for high tuition costs. Given the widespread resistance to further tuition increases and the competition between the legitimate goals of tuition remission and research, it is unlikely that substantial additional resources for basic work in the life sciences will come from the research universities themselves.

At a typical private research university, only about 85% of the indirect costs of sponsored research has been recovered in recent years. The situation in public research universities is probably no different. The shortfalls result from the fact that many government agencies, as well as many private foundations and corporations, have refused as a matter of policy to pay full indirect costs for research. Other agencies, which negotiate indirect costs according to some formula, have required "cost-sharing" by the university; have refused to accept outside-the-formula "special studies", which justify above-average costs; or have placed non-negotiable "caps" on particular items in the indirect-cost pool, generally for the explicit purpose of limiting outlays for research grants. As budget-balancing continues to occupy center stage in Congress, research universities face a likely decline in their real levels of federal support.

To maintain an adequate volume of research and the infrastructure to support it (the object of indirect-cost recovery), research universities must find alternative sources for research funding. Although increased gift income is one possibility, undergraduate financial aid and research will probably continue to compete with one another for scarce tuition dollars, at least at private

research universities. Successful efforts to maintain levels of research support will probably lead to fewer low-income students at these institutions. Alternatively, maintaining current levels of financial aid and student diversity will mean less internal support for research. Only if universities can achieve substantial cuts in other areas of costs can this tradeoff be avoided.

One of the principal components of a university's budget is faculty salary, there is a natural administrative interest in opportunities for savings in this line. Unfortunately for this purpose, the abolition of mandatory retirement at a designated age has narrowed one such opportunity: it appears that a substantial number of professors are choosing to retire at later ages. Even a modest increase of 3–5 years in age of retirement (to 68 or 70, instead of 65) will mean an increase of 10–15% in the mean duration of a faculty career and an equivalent decrease in the number of people who can enter that career, all other things being equal. That not only slows the rate of faculty replacement, but it increases salary costs because senior faculty tend to be more expensive than their younger colleagues. It is not yet clear what strategies might help to reverse this trend. Attempts by universities to do so, by offering incentives to retire, do not appear to have saved money in the short run.

The current faculty age distributions at almost all colleges and universities virtually guarantee that the coming years will see vacancies that can be filled by younger scientists. The situation does not, however, guarantee that there will be vacancies for research-oriented faculty, nor that the positions available will be tenure-track. Universities seem to be responding to financial pressures by hiring more nontenure and part-time faculty. The reduction in tenure-track opportunities might make academic research posts less attractive to young scientists and have an impact on the extent to which talented college students are drawn into life-science research.

Medical schools, which are generally parts of research universities, now face additional problems in maintaining a healthy research environment. Academic health centers (AHCs) include basic-research faculty and clinical researchers, as well as medical educators and physicians; these scientists work collectively to provide teaching, research, and clinical care. AHCs emerged during the period of unprecedented growth in the health-care sector that followed World War II. Substantial resources became available for building health-care partnerships among medical schools, university hospitals, and private medical centers. The resulting AHCs deliver multiple health-care services.

AHCs have flourished on federal dollars, along with a steady stream of income from faculty practice plans. Indeed, some AHCs today receive over 50% of their income from revenues for patient care. Faculty practice plans in 1993 provided at least $2.4 billion in support of academic programs, including undergraduate medical education ($702 million), graduate medical education ($594 million), and other academic support ($244 million) (Jones and Sanderson 1996). Faculty research grants also provide income to AHCs in the form of faculty and staff salary support and indirect-cost recovery. However, shortfalls in indirect-cost recovery and the requirement of some sponsors for cost-sharing create a financial burden for the recipients of the funds. Such financial losses are generally compensated for by the gains in intellectual capital that result from greater scientific sophistication, increased academic prestige, more numerous publications, and

sometimes patents, which can produce additional income. In sum, research in most AHCs is heavily subsidized by clinical income, which is vulnerable to policies that reduce the revenue from patient care.

The research mission of AHCs has contributed significantly to America's preeminence in medicine and biomedical science, but the landscape is changing fast, and the future of research at AHCs is, at best, uncertain. Radical change occurred in 1990 when managed care started to replace the medical faculty's traditional fee-for-service operation; competition from health-maintenance organizations for patients now threatens income flow to AHCs. AHC administrators are scrambling to reorganize their hospital and clinical services and are attempting to establish their own networks of clinical specialists to compete in the primary-care market.

Mergers, acquisitions, and joint ventures with various health-care providers are now common. Such maneuvers are accomplished, however, at the expense of specialty care and of graduate medical education.

It is not yet clear how the new arrangements will affect biomedical research and education, which principally have been conducted by doctors whose salaries were partly subsidized from patient-care income. More than ever, the faculty engaged in research will be expected to fund most, if not all, of their salary, as well as their laboratory costs, from their own research grants. This change is coming at a time when grants are harder than ever to get. In some AHCs, the basic-research enterprise is already being reduced as faculty leave or retire. One can reasonably expect the current stringent conditions will shrink the research enterprise at most AHCs. Moreover, the net impact of managed care is likely to be a devaluation of research success as a criterion for promotion and reward in most medical schools. Without cutting-edge research and a strong academic environment, progress in medical research could languish. It appears that the remarkable era of the traditional AHC is ending, but the full impact of this sea change on the management, philosophy, and morale of medical-school faculties has yet to be realized.

At the same time, financial support of research from pharmaceutical companies has increased substantially in recent years and makes up some part of the support lost because of changes in clinical-practice income.

Public policies affecting agriculture and forestry were designed to enhance the productivity of US farms and forests. They were focused in particular to enhance the economic status of farmers and to promote general public welfare. The land-grant university system, with its strong components of experiment station research and extension service, has nurtured an agricultural enterprise that allows the American public to spend a lower percentage of its income for the purchase of food than any other country in the world: between 1956 and 1996, field-crop yields have about tripled while the acreage devoted to agriculture has decreased. The US agricultural research enterprise is therefore perceived by most people to be a bargain.

Over the last 30 years, there has been a serious change in the support of agricultural research. Between 1960 and 1990, the estimated funding for private research in agriculture has tripled; it currently exceeds the investment by both state and federal agencies. These funds have come from chemical, petroleum, and pharmaceutical companies, and a large percentage involves venture capital for biotechnology investments. Although the record of expenditures by companies is not fully disclosed, the sum

probably now exceeds $3.5 billion per year. As private investments have increased, there have been major shifts in the kinds of research that are funded. Support of plant breeding has quadrupled and that of animal health has tripled while funds for research on machinery have declined from 36% to 12% of the total invested.

Investments by the states in agricultural research have continued to increase; in sum, they are now much higher than the corresponding federal appropriations. Indeed, the rate of increase in federal support has not kept pace with the needs of teaching institutions. The result has been indirect but negative: a decline in the number of instructional positions that are directly related to agriculture. Many land-grant universities have established programs in molecular biology, biotechnology, sustainable or alternative agriculture, and environmental sciences. Additional changes have been made at some universities to integrate forestry and agricultural research programs, emphasizing studies on regional ecosystems and landscape and wildlife management research programs. The cadre of applied ecologists will need to be increased to cope with these changes in research perspectives.

There is now a pressing need for agricultural-research biologists who are responsive to changing societal requirements to insure the continued availability of agricultural products at a relatively low cost to the consumer while maintaining economic stability for the growers. Such scientists will be essential if we are to provide areas for recreation and ecologic diversity, to conserve and restore damaged ecosystems, and to reduce our dependence on pesticides and other chemicals. Moreover, there will be an ever-increasing need for biologists capable of using the major advances in molecular biology to increase the availability, quality, and safety of food under circumstances that will ensure the sustainability of agriculture and natural resources. The situation suggests that more, not less, should be invested in the agricultural life sciences, broadly defined. The current heavy reliance on funding from the private sector carries some danger that some basic-research problems with less potential for commercial payoff will not get the attention that they need and deserve. That is already evident in the decline of support by major agricultural-chemical companies of research on microbiologic control agents for plant diseases. The emphasis of these companies is on research on and development of transgenic cultivars with disease and insect resistance.

Before the post-World War II burst of federal funding that created the research-intensive, PhD-granting university, industry was the major supporter of life-science research, and PhDs regularly entered industrial careers. Some mentors and trainees today believe that the only respectable career aspiration is academic research. That opinion is sharply out of phase with the fact that only one-third of PhDs currently obtain academic research positions, whereas jobs in industry have increasingly provided career opportunities for life scientists.

Chapter 3 shows that during the 1980s, when the number of academic research positions was no longer growing rapidly, industry became a major source of jobs in the life sciences. The trends in the 1990s suggest, however, that the growth in the number of industrial research positions might not be as robust in the future as it was in the early 1980s. Several features of industrial organization and patterns of employment are affecting the availability of careers in the life sciences, as discussed briefly below.

The number of jobs for doctoral-level

microbiologists is projected to grow at an annual rate of 6%; about 15% of the growth represents hiring of postdoctoral fellows, not scientists with permanent positions, according to a recently completed survey by the American Society for Microbiology (Van Ryzin and others 1996). The ASM survey showed, however, that the fastest growth was in emerging fields of biotechnology, such as bioremediation, molecular immunology, and antimicrobial chemotherapy. For some pharmaceutical companies, the highest level of new hiring is in such fields as drug formulation. Chemistry and toxicology show a steady rate of hiring that primarily reflects attrition, with few new positions appearing. By comparison, fields like molecular biology, which saw strong growth in the middle 1980s, are showing no further growth in the 1990s, and replacement hiring might shift toward other life-science disciplines. The ASM survey showed that 57% of industrial respondents forecast increased hiring, but these companies also told the surveyors that future employees must be more flexible and less specialized than their predecessors. At one leading pharmaceutical firm, an increasing number of open positions that were once filled with scientists trained at the bachelor's and master's level are being refilled with PhD scientists.

In the pharmaceutical and biotechnology industries, the late 1980s and 1990s saw a steady stream of consolidations that resulted in substantial corporate savings with a concomitant disappearance of research positions. The large number of experienced researchers who are therefore on the job market has made it difficult for new PhDs to compete for open positions. In addition, many activities that used to consume large amounts of research time (such as peptide and oligonucleotide synthesis, protein and nucleic acid sequencing, monoclonal and polyclonal antibody production, and receptor-binding assays and immunoassays) have become sufficiently routine that robotics and automation are useful options. Further efficiencies of scale have come from the emergence of new companies that provide the services to pharmaceutical and biotechnology enterprises, but the new positions at these service companies simply offset some positions lost elsewhere in industrial research.

During the rise of biotechnology in the 1980s, fundamental research was a major part of the work being done by the scientists in the new positions. However, the nature of the industrial research positions has now shifted. The emphasis is now on transforming the fundamental discoveries of the 1970s and 1980s into commercial uses and applications.

Industry continues to down-size, consolidate, and become more efficient. The total volume of industrial research will probably continue to increase but this research is for the most part focused on applied research that has short-term commercial payoffs. Moreover, research on agriculture-related topics is constrained by the commercial value of discoveries. Unlike products with commercial medical applications—whose cost has not, until recently, been prohibitive to development—agricultural research for commercial development is often constrained by the cost of the potential products. Consumers are not willing to pay as much for agricultural innovation as they have been for medical advances; the kind of research that can profitably be pursued in the commercial sector of agricultural research has thereby been constrained.

Even when one understands the economics of a given branch of industrial science, it is generally hard to use the knowledge to predict where increased workforce needs will emerge. Very few people predicted the dramatic emergence of biotechnology before the 1980s.

New fields of industrial research that increase the demand for life-science researchers might emerge. It must be remembered, though, that just as automation and increased efficiency have come along in biotechnology research (for example, in DNA sequencing), technologic innovations that substantially reduce the demand for PhD researchers can be expected to change the patterns of employment of newly trained life scientists.

As pointed out in chapter 3, the overall fraction of recent PhDs who are employed in government is decreasing, particularly in the older cohorts. If current trends toward government down-sizing and budget balancing continue, federal employment of research scientists cannot be expected to increase. Some growth can be expected, though, in selected fields that are not research-intensive. For example, as reported by Katterman (1996), the number of biotechnology-patent applications filed in the United States has grown about 10% per year since 1990. As more and more genetically engineered products near the marketplace, there will probably be new employment opportunities for life-science PhDs in federal patent-licensing offices and in some regulatory agencies, such as the Food and Drug Administration.

Life-science PhDs who seek academic careers with a greater emphasis on teaching might find satisfying careers at several kinds of non-PhD-granting institutions: conventional 4-year liberal-arts colleges that award bachelor's and sometimes master's degrees, 2-year junior and community colleges whose degree is usually an associate in arts, and public and private elementary and secondary schools. An analysis of current employment patterns shows that PhDs are more likely to be found in the 4-year colleges, less likely in community colleges, and comparatively rarely (but not totally absent) on secondary-school science faculties. As the present crop of life-science PhDs in postdoctoral positions seek more permanent jobs, these employment patterns might change, so it is important to examine the current situation with some care.

About 20% of the life scientists who are tenured or on the tenure track are now teaching at the roughly 1,150 4-year colleges or universities that do not offer the PhD. These institutions have grown greatly over the last 3 decades, and they have been an important source of employment for recent PhD recipients. Unlike the situation at PhD-granting institutions, the number of faculty positions at 4-year non-PhD-granting institutions has continued to rise, and the number of positions held by life scientists within 10 years of receipt of the PhD increased in both 1993 and 1995 after a period of decline. Because of high student interest in biology as a major, as well as the common focus on preparation for medical school, many life-science departments have grown over the last decade; this trend might continue as students who make up the ''echo" of the baby boom matriculate in college. The US Department of Education projects an increase of 0.7 million students in 4-year institutions during the next decade. Assuming that teacher:student ratios remain constant and that there are no changes in instructional practices that might diminish labor requirements, these trends could lead to an increase in the number of life-science faculty.

Most of the biology departments in these colleges are staffed by PhDs who are well trained

in research, and most of the faculty are expected to conduct research that employs and trains students. The leading liberal-arts institutions are well known as the source of some of the best graduate students at the top research universities, and it is the research opportunities that they had as undergraduates that prepared these students so well for graduate education. A few such institutions also offer the master's degree. Faculty members have opportunities to pursue their own research interests, but most liberal-arts college professors still spend the majority of their working time instructing students. Salaries at liberal-arts colleges are on the average near or only slightly below those at research universities, but the best-paid teachers at these 4-year institutions are better compensated than those at low-paying universities.

Because most life-science PhDs and postdoctoral fellows have concentrated intensively on research, they have comparatively little experience in teaching, and their qualifications might not be attractive to teaching-intensive colleges. Some graduate students can take advantage of new programs at a number of PhD-granting institutions that offer students exposure to teaching in a more rigorous manner. A small number of "teaching postdoctoral fellowships" have also been developed. One such program (funded by a private foundation) was described to the committee at its public hearing; it provides postdoctoral trainees with 2 years of teaching experience supervised by a mentor. Such a program seems likely to be effective in preparing participants for positions at teaching-intensive institutions.

The committee found that the 1,471 institutions at this level of higher education employ only about 600 PhDs in life sciences, and the prospects for substantially increasing this number appear to be small. There might be an increased demand during the coming decade, fueled again by the echo generation of the baby boom, which is predicted to increase enrollment at 2-year colleges by about 11%. The impact will probably be quite selective, in that it is apparent that many, perhaps most, of the 2-year institutions do not have a PhD in the life sciences among their faculties.

Hiring projections in the COSEPUP report (COSEPUP 1995) suggested that the echo of the baby boom could lead to numerous new positions for K-12 teachers, providing alternative career opportunities for science and engineering PhDs. Our committee believes that this change will probably create a demand for PhDs only at the secondary-school level, and even here the demand is likely to be small. About 0.5% of PhDs in the life sciences are currently K-12 teachers. At that rate, one might expect that 35–40 of the roughly 7,500 PhD's graduating per year would enter precollege teaching. If the rate of entry into secondary schools triples owing to increases in the student populations and increased enthusiasm for the life sciences, the number of PhD life scientists that could be absorbed would be only somewhat more than 100 per year. That is less the 2% of the current production of life-science PhDs so this source of jobs is not likely to have a major impact on career patterns for life scientists.

There are, furthermore, obstacles to the employment of PhD scientists in secondary schools, notably the low salaries and the teacher-certification requirements. Although pay scales for secondary teachers with PhDs are normally higher than for teachers with bachelor's or master's degrees, they are generally lower than the salaries for entry-level assistant professors. Scientists at the end of a 5–12 year period of postbaccalaureate training might well regard secondary-school teaching as a bad bargain. In addition, most states require credentials for a

teaching certificate that would necessitate a year or more of additional training in education—also an unappealing lengthening of prejob training. Although a few states have special programs to train candidates with advanced degrees for public-school teaching, the burdens of supporting oneself and paying for this additional training are likely to be serious disincentives. Finally, experienced administrators in public-school systems have offered the opinion that life scientists who are extensively trained in cutting-edge research would not find school teaching captivating.

With the increase in biotechnology patents and an upsurge in the use of molecular biology as a tool in criminal investigation there has been an increase in the opportunities for life-science PhDs to enter the legal profession. The patent field appears to be dominated by about a dozen large and medium-sized firms. Estimates made in 1997 by patent lawyers at two of those institutions indicate that 20–100 new jobs would become available per year for life-science PhDs. It is customary for PhDs who begin working at law firms to go to law school at night for 3–4 years to earn the law degree that is deemed a necessary credential. Some large firms have clerkship programs that cover law-school costs in exchange for a commitment to continue working for the firms. There is a recent trend to hire PhDs, rather than master's-level scientists, for these jobs because of the large number of highly qualified candidates. PhDs also add to a firm's reputation.

There is a growing interest in journalism among life-science PhDs. Such opportunities appear to be largely associated with the numerous scientific journals that are published, rather than with the more limited number of publishers who handle scientific books. A few life-science PhDs currently working in publishing whom we spoke with thought that future opportunities in the field would probably be constant or perhaps increase slightly. However, competition for careers in journalism is often high. For example, one journalist with a recent PhD in life science moved from a highly regarded specialty journal to a more general publication. There were about 200 applicants for the latter position, and about 50 applications were received for the position vacated at the more specialized journal. Not all the applicants were PhDs, but a doctorate and journalistic experience would appear to have provided the best credentials. The Internet was cited as a medium with particularly good growth potential for scientific journalism.

Some life scientists find positions with private foundations and various other scientific concerns. Again, the competition for such positions is steep. A former assistant professor in the life sciences reported to the committee that there were more than 200 applicants for her present position managing the research-grants program of a philanthropic organization. That figure and others mentioned earlier indicate that there is considerable interest in nontraditional career paths among life scientists. Most PhD programs do not, however, offer the broader exposure and training that would be helpful for entering nontraditional career. The question of whether life-sciences PhD programs should change to offer this additional training is addressed in chapter 6.

In summary, our findings suggest that the number of positions in nontraditional fields of employment for life-science PhDs appears to be rather small, and that the competition for these jobs is strong. The committee acknowledges that it cannot predict the emergence of entirely new employment opportunities that might change employment characteristics considerably. Several sites on the World Wide Web (for example, Next Wave: An Electronic Network for Young Scientists) offer career information that might be

of interest, and appendix G contains a list of Web sites that provide data and career information for life scientists.

Clotfelter CT. 1996. Cost escalation in elite higher education. National Bureau of Economic Research monograph. Princeton, NJ: Princeton University Press .

COSEPUP (National Academy of Sciences, Committee on Science, Engineering, and Public Policy). 1995. Reshaping the graduate education of scientists and engineers. Washington, DC: National Academy Press

Jones R and Sanderson S. 1996. Clinical revenue used to support the academic mission of medical schools: 1992–1993. Academic Med 71:3

Katterman L. 1996. Biotechnology patent boom offers career opportunities for scientists. The Scientist 10:15-6.

Ruzek JY, O'Neil EO, Williard RL, Rimel RW. 1996. Trends in US funding for biomedical research. San Francisco: UCSF Center for the Health Professions.

Van Ryzin G, Dietz S, Weiner J, Wright D. 1995. The employment outlook in the microbiological sciences, 1995. http://www.asmusa.org/pasrc/empoutlk.pdf.