Read an Excerpt
Prospects for Faculty in the Arts and Sciences
A Study of Factors Affecting Demand and Supply, 1987 to 2012
By William G. Bowen, Julie Ann Sosa
PRINCETON UNIVERSITY PRESSCopyright © 1989 Princeton University Press
All rights reserved.
American higher education has experienced periods of rapid expansion and retrenchment since World War II, with swings that have been sharp and at times destabilizing. The war itself had major consequences: returning G.I.'s swelled enrollments, and there was a new appreciation of the power of scientific research. After a brief hiatus in the early 1950s, the launching of Sputnik in 1957 stimulated another strong wave of interest in higher education. Then, in the 1960s, the baby boom combined with rising enrollment rates to create one of the greatest expansions ever seen. Shortages of faculty were widely proclaimed, faculty salaries increased markedly, and there was talk of a "golden age" for academics. Both students and graduate schools responded enthusiastically to the heightened demand for faculty, and the number of doctorates awarded increased more than threefold between 1958 and 1970.
An unintended result was the creation of a substantial excess supply of potential faculty members in the 1970s, when higher education entered what came to be called a "new depression." The number of doctorates awarded subsequently declined almost as rapidly as it had risen, creating a roller-coaster pattern in many fields of graduate study. In recent years, improved economic conditions have permitted at least a modest recovery in the financial circumstances of faculty members, but there has not yet been any significant change in the number of new doctorates awarded annually.
We are now in an unsettled period for higher education, and even short-term prospects for faculty members are far from clear. The current climate of thinking is illustrated by a recent article entitled "Uncertainty is Rampant as Colleges Begin to Brace for Faculty Shortage[s]. ... The title is apt, both in indicating a growing concern about faculty staffing problems and in acknowledging the extent to which people are unclear about what lies ahead and what steps should be taken to ward off possible dangers. At the same time, it is widely agreed that the future shape of higher education will be affected profoundly by decisions that are made — and not made — over the next few years.
When the presidents of a number of colleges and universities were asked by the Andrew W. Mellon Foundation in the spring of 1988 to identify pressing problems facing higher education, they were nearly unanimous in expressing deeply felt worries about their ability to recruit outstanding faculty in future years. Yet these presidents were also quick to acknowledge that most of their expressions of concern were based on either knowledge of local conditions or anecdotal information.
At present there simply is no body of evidence and analysis that would allow a comprehensive assessment of the outlook for faculty staffing. That reality — the need for a framework shaped by reasonably "hard" data — has been a primary motivation for our research.
The purpose of this study, then, is to develop a systematic analysis of the factors affecting prospects for faculty in the arts and sciences over the next twenty-five years. We hope to provide both a clearer sense of whether the "boom" and "bust" pattern of faculty staffing is likely to repeat itself and an improved understanding of how to avoid such disruptive and inefficient cycles.
We are well aware that this particular study is a highly quantitative analysis of issues with personal, political, and social dimensions that reach far beyond projections of numbers of students and faculty members. At stake is the quality of teaching and research, by which we mean the ability of faculty members to encourage active engagement with the kinds of learning and scholarship that enable colleges and universities to be the source of new ideas as well as effective communicators of what is already known. Also at stake is the ability of higher education to continue to be an avenue of opportunity for individuals from diverse backgrounds.
The continuing capacity of our colleges and universities to serve such large purposes depends on the willingness of society at large, and especially institutions of higher education, to address some very direct questions. What incentives should be provided to encourage aspiring academics? What should universities do to strengthen their graduate programs? What is the right role for the federal government in this area, given the widespread agreement that a highly educated populace constitutes one of this country's major sources of comparative advantage?
There is, we believe, a broadly based willingness to think hard about such questions. The acrimonious nature of much of the recent debate over educational philosophy and curricula (the role of the "great books," for instance), and over levels of tuition and the cost of higher education, has masked the degree to which people on all sides of these issues are committed to the educational enterprise itself. A widely shared dedication to higher education is special to this country, even if it sometimes appears to be taken for granted.
While this study adopts the usual convention of analyzing factors that affect the demand for and the supply of faculty, it differs from its predecessors in significant respects. One distinguishing characteristic is that we have made determined efforts to look separately at major fields of study and at principal sectors of higher education. Prospects for the sciences, for example, may differ from prospects for the humanities. Liberal arts colleges may need to see their future in terms other than those which are most relevant to the large universities.
This approach can be thought of as "disaggregation," but it is better understood as "aggregation" in that we build results for the arts and sciences as a whole from data for particular fields and sectors of higher education. This emphasis has made our research more complicated than it would have been otherwise, but we think that the resulting increase in the precision of the results, and the attendant ability to compare changes across fields of study and sectors of higher education, justify the effort.
In designing our research, we sought to construct a "machine" that could be used to generate a variety of projections of supply and demand. This involved identifying clearly both the relevant "parts" and the ways in which they interact. We have tried equally hard to be sure that each part is congruent with all the others (covering the same universe of faculty members, fields, institutions, time periods, and so on). This permits ready substitution of alternative assumptions and makes it possible to test the sensitivity of results to such substitutions. As a consequence, we are able to learn that some components of the machine, such as student/faculty ratios, are powerful in determining outcomes. Other factors, such as retirement patterns, can be seen to be relatively unimportant quantitatively, even though they may be very important for other reasons.
While we make use of supply and demand terminology, we wish to emphasize that this is in no sense an econometric analysis. Throughout most of the pages that follow, we use the terms "demand" and "supply" in the crude sense of the expected number of positions to be filled and the expected number of candidates for vacant positions — assuming specified salaries, terms of employment, and other labor market conditions. In short, we are concerned primarily with shifts in demand and supply functions over time, not with the shapes of demand and supply curves seen as functions of salaries and other conditions of employment at points in time. Although we discuss adjustment mechanisms, including both demand-side and supply-side responses to changing labor market conditions, "price" variables as such do not play a major role in our effort to project longer-term trends.
We also wish to emphasize that the projections of labor market conditions presented in this book are designed primarily to quantify the consequences of adopting various assumptions about the future. Properly understood, they can serve several worthwhile ends. They provide an "early warning system" that may allow us to anticipate emerging problems in time to take some sensible corrective actions. In addition, they can help inform both the career choices of individuals and discussions of policy within academic institutions and governmental bodies. In short, they can assist in testing out the implications of various courses of action. But such projections should never be thought of as predictions of what will actually happen or as instructions as to what should happen.
One of the most perceptive commentaries on the value of the kinds of projections we have tried to develop was written in the mid-1970s by Allan Cartter in a little-known article that preceded his major study. In reflecting on some of his earlier efforts to study the supply and demand relationships for teachers, Cartter observed (1974, pp. 282–83):
One should draw a careful distinction between projections and predictions; the former may illustrate the consequences of current trends and thus serve to alter the course of events. In a meaningful sense, successful projections may be those that turn out to be poor predictions of actual events. My projections of 1964 and 1965 that the academic labor market would reverse itself in 1969 and 1970 and become a surplus labor market could be called unsuccessful in the sense that few persons took them seriously, thus permitting them to become accurate predictions.
Any set of projections that uses fixed coefficients is suspect because the coefficients themselves are likely to change in response to variations in labor market conditions. Ideally, we would use a more sophisticated mode of analysis that included well-specified behavioral relationships and that took account of feedback mechanisms and the recursive nature of some of the implicit labor market equations. Perhaps subsequent studies will incorporate improved techniques of this kind. In the interim, we have tried to indicate the general nature and significance of such relationships and interactions, especially when we integrate different kinds of assumptions (about enrollment patterns and student/faculty ratios, for example) and when we discuss adjustment mechanisms.
We agree with Cartter that there is a place for studies that rely on projections as well as for studies that are more behavioral in their approach. In at least a limited sense, this study represents an effort to take some advantage of both ways of thinking. We have made heavy use of the kinds of simplifying assumptions that facilitate longer-term projections; and we have also sought (especially in the last part of the book) to recognize the ways in which the responses of individuals and institutions, and the workings of markets, will affect what actually transpires. In adopting this dual approach, we are aware that, as Isaiah Berlin once wrote (1978) in an entirely different context, "the middle ground is a notoriously exposed, dangerous, and ungrateful position."
In pursuing this general approach, we have chosen, first, to define "faculty" quite carefully. We limited our universe to those in the professorial ranks plus instructors and lecturers (excluding "adjunct faculty"). We also elected to focus on full-time faculty. Failure to distinguish full-time faculty from part-time faculty would have made it more difficult to come to clear conclusions — pertaining, for instance, to changes in age distribution — that otherwise might be masked by the shifting weights of full-time versus part-time faculty. Our universe of faculty also has been limited to those holding doctorates. This decision, driven in part by the nature of one of our primary sources of data (the biennial Survey of Doctoral Recipients), is consistent with our general desire to look closely at the core group of full-time faculty.
In addition, our study focuses almost entirely on faculty members in the arts and sciences. The decision to exclude faculty members in engineering, agricultural sciences, medical sciences, education and other professional fields in no way reflects on the importance of these subjects. Their circumstances are often quite different from the circumstances confronting the arts and sciences, and simply combining all schools and disciplines seemed to us to entail the serious risk of confusing the analysis. Also, the centrality of the arts and sciences to all of higher education justifies, we believe, some concentration of attention on these disciplines.
Within the arts and sciences, we have paid separate attention to eight broad fields: (1) mathematics; (2) physics and astronomy; (3) chemistry; (4) earth, environmental, and marine sciences (hereafter referred to as "earth sciences"); (5) biological sciences; (6) psychology; (7) social sciences; and (8) humanities. In addition, some of the subsequent analysis includes a breakdown of the humanities into four components: (a) history; (b) English and American languages and literatures; (c) foreign languages and literatures; and (d) other humanities. We have also found it useful in parts of the analysis to work with three "clusters" of fields: (1) mathematics and physical sciences; (2) biological sciences and psychology; and (3) humanities and social sciences.
We have also devoted a great deal of attention to distinguishing among sectors of higher education. Specifically, we have used the institutional classification system of the Carnegie Foundation for the Advancement of Teaching, but we have elected to combine some of the categories and to concentrate on five principal sectors, which we have defined as follows: (1) Research I; (2) Other Research and Doctorate; (3) Comprehensive I; (4) Liberal Arts I; and (5) Other Four-Year. (See Appendix A for a full discussion of the ways in which we have combined types of institutions.) In general, it was not possible to provide separate breakdowns for public and private institutions. However, we do comment briefly (especially in Chapters Six and Eight) on some of the major differences associated with the public/ private classification.
Over at least the last two decades, there has been strong concern about the composition of faculties classified by gender and race. Extremely important questions of educational and social policy grow directly from studies of the participation of women and of minority groups in higher education. However, these questions — important as they are — are not the subject of this study. To have included cross-tabulations by gender and by race would have produced an overwhelming mass of statistics concerning subjects that deserve separate and intensive analysis.
The period covered by this study extends from the present through 2012. We recognize that a twenty-five-year time-span is very long, and we certainly do not believe that anyone can foresee many of the most important changes that are sure to occur between now and 2012. But some of the major demographic factors affecting higher education do cast long shadows, and it seems desirable to look as far ahead as possible. It is also the case that many of the policy questions that must be considered require a long time horizon, since changes can often be introduced only slowly, and long response lags need to be taken into account. This is true, for example, of changes in the distribution of faculty by field and in the number of recipients of doctorates.
We are all too aware that we have had to make what can generously be called "bold" assumptions at various points in the analysis. In a number of instances other assumptions might well have been made, and we therefore have tried throughout to explain quite precisely what has been assumed, to provide derivations of important factors used in adjusting data and making projections, and to indicate how sensitive our results are to the particular assumptions chosen. In these ways, we have tried to avoid claiming too much, and we have also tried to make it possible for others to substitute their own assumptions for ours. At times this approach may make for somewhat heavy going, but it has seemed the only responsible way to proceed.
Another of our objectives has been to identify and highlight opportunities for further research. Some of the trends we describe are parts of larger changes in society that are worthy of more comment than is appropriate here. The broad framework that we have sought to provide may also stimulate case studies and other efforts to probe particular patterns of institutional behavior that we did not investigate.
There is considerably more to be said about the demand side of the labor market equation than about the supply side, and that is where we begin our analysis. It has become customary to divide the demand for faculty members into a replacement component and an enrollment-driven component. This is, however, a somewhat arbitrary distinction. Decisions made by institutions concerning the creation or elimination of faculty positions are often affected by the interplay of anticipated retirements (or departures for other reasons) and increases or decreases in enrollments.
While the number of faculty positions assigned to an academic department depends on enrollment, other factors are involved as well: the perceived "critical mass" required if a department is to function effectively; the faculty time to be devoted to research and to other functions, such as governance and outreach activities; opportunistic considerations (the sudden availability of an outstanding candidate for appointment); and the financial resources that are available. All of these considerations are reflected in the student/faculty ratio.
Excerpted from Prospects for Faculty in the Arts and Sciences by William G. Bowen, Julie Ann Sosa. Copyright © 1989 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.