From Chronicle of Higher Education:
This year marked the 25th anniversary of the most infamous academic-labor study of all time, "Prospects for the Faculty in Arts and Sciences." The study, led by William Bowen, then president of Princeton University, set itself the task of projecting "demand and supply" for faculty a full quarter-century into the future—forecasting the so-called job market right up into our present decade.
Contrary to the widespread knowledge of permanent retrenchment and adjunctification, the study projected that a huge "undersupply" of people holding doctoral degrees would manifest by 1997. However, nothing of the kind transpired. In reality, the perma-temping of the faculty continued on the same steeply upward trend line as before.
The Bowen study’s misreading of the future raises two questions. What was wrong with the assumptions guiding it? And why did an effort with so many flaws receive such an uncritical greeting? The answers remain surprisingly relevant.
The Achilles’ heel of the study and of similar efforts still published by professional groups like the Modern Language Association and American Historical Association is their mistaken approach to the "demand" side of the equation. If at any point from 1989 to the present you wanted to publish a valid "labor market" analysis of higher-education research and teaching, you would have to begin with the acknowledgment that if it is a market, it is one with steadily increasing "demand" for labor, yes, but contingent labor, not traditional tenure-track jobs.
By 1989 the academic labor system was already two decades into "restructuring" the professorial job. All the evidence then available (data from the National Center for Education Statistics, testimony of job seekers, education-management literature, critical literature, budgeting and workload information, program growth) affirmed that most campus administrations were moving as much work as possible out of professorial jobs.
Since the late 1960s, under the rubric of "retrenchment," administrations had been reassigning the labor previously performed by tenure-track faculty to staff members, postdocs, contingent employees, and graduate fellows. Even undergraduates were already being mobilized to do work previously performed by faculty: grading, tutoring, mentoring, and monitoring of residence halls.
Against all of this evidence, Bowen assumed that the move toward contingent hiring was temporary. Explicit in his projections was the belief that when campus administrations could afford to hire tenure-track faculty, they would do so preferentially.