the minnesota review n.s. 50-51 (1999)Cynthia GabrielGraduates as Students, Graduates as Employees: Labor Unions in the AcademyIn December 1998, more than 5,000 teachers walked out of University of California classrooms to shoulder picket signs. Their one demand was union recognition. This end-of-term teaching and grading strike was the culmination of fifteen years of organizing. Notable because it coordinated a diverse population on eight campuses, this strike was carefully planned for more than two years. Union elections in the spring of 1999 on all eight University of California teaching campuses are an unprecendented victory for the graduate employees of UC and for the entire student labor moment. The last decade has witnessed a sharp rise in union organizing among teaching assistants, readers, tutors, research assistants, and teaching fellows in major U.S. universities. The first graduate student employee union was recognized at University of Wisconsin more than thirty years ago, in 1966; since then, thirteen have followed suit. From 1990 to 1996 one new graduate student employee union was recognized every year. More than twenty-five are seeking recognition now; eight of those are universities within the University of California system. Strikes and other job actions by these workers affect academic institutions across the United States, from Yale to the University of Iowa to UCLA. The University of California unions, affiliated with the UAW, have now declared victory in their fifteen-year struggle for union recognition. UCLA was the first campus to have state-ordered elections in March 1999. This election was won by a decisive 718-269 or 73% vote. Seven other UC campuses will voted in April and May. The recognition issue for graduate teachers seems decided at University of California once and for all. University administrators across the nation have looked to UC for guidance and precedence on this labor issue. Until this spring, UC's example was straight from an 1880's industrial employer's playbook: stall, stall, divide the workforce, and stall some more. Stalling was accomplished primarily through lawsuits; time and again, during the past fifteen years, UC administrators said, "Just wait. Wait for the courts to decide." In the latest case to determine whether graduate student employees had collective bargaining rights, the Public Employment Relations Board (PERB) had to consider "Prong Two" of the Higher Education Employer-Employee Relations Act.i Prong Two requires PERB to weigh the benefit graduate students receive as students against the benefit the University of California receives as the employer. The question to be decided was which entity benefited more. The test campus was UCLA. Even in the absence of a PERB ruling, however, the University could recognize the unions. Graduate employees argued that this PERB case and the Prong Two test were simply stalling tactics. The University had been informed by PERB that a majority of employees on all eight campuses had signed union cards and could be recognized. In December 1998, what was likely the largest U.S. academic personnel strike was called to demand union recognition. Four days into the strike, senior members of the California state legislature called for a 45-day cooling off period. For the first time, UC administrators agreed to meet with the unions face-to-face to discuss union recognition. During the cooling off period and quite coincidentally, PERB issued its final ruling on the UCLA case. PERB found that teaching assistants, teaching fellows, readers and tutors had collective bargaining rights. They ordered a representation election at UCLA for March 9-11, 1999. The remaining seven campuses will hold elections later in the spring. Why are union drives happening now? Why are they winning? This paper will answer these questions through a historical analysis of who has made up the category "graduate student" and how changing teaching patterns at U.S. universities have impacted the lives and experiences of graduate students. First I will examine who graduate students were and who they are now. I will point out the ways that graduate students today differ from their "ancestors" in terms of age, gender and financial resources. I will use this demographic data to show that socio-economic support systems and the economic expectations of graduate students have changed over time so that more and more graduates experience teaching assistantships as jobs, rather than apprenticeships. I will also show how these changes have disproportionately affected white women and certain students of color. Secondly, I will analyze the experience of graduate school itself. How does employment (as a teaching assistant, for instance) fit into graduate education? How have funding patterns changed or stayed constant over time? How have the shifting tenure and employment patterns of higher education institutions impacted the experience and expectations of graduate students? As a union organizer and graduate student, I see activism in the academy everywhere I look. This paper is an attempt to see activism and academic work as potentially intertwined and interdependent, and to turn my activist and academic gaze on the institution, the University of California, in which I find myself. PART ONE: HISTORY OF Ph.D.sOver the fifteen-year period that University of California has fought unionization, it has argued vehemently that graduate student employees are students, not employees. Therefore, it doesn't matter if a majority have signed union cards or voted to have union representation. Graduate teachers do not have the same rights as other U.S. employees because they are not employees. This argument is so important to the University of California that it has spent an estimated $3 million in the last three years to fight its graduate student employees in court. In the fall of 1998, the University of California took the further step of breaking the law in order to maintain the stance that TAs are not employees (October, 1998 letter from President Atkinson).ii This view of graduate students arises out of certain historical conditions which do not pertain today—at least not as intensely. Though I do not mean to suggest that graduate student employees of previous generations were any less deserving of collective bargaining rights, graduate work has changed in character and form since 1863. Ph.D. vs. Professional and Master's DegreesFirst of all, let me clarify that I am speaking primarily of Ph.D. graduate work and my arguments are less relevant to master's degree students or professional school students. In fact, I argue that misunderstandings of the word "graduate student" have much to do with legal, media and popular representations of the union struggles of academic student employees (ASEs). Because only 1% of the population receives a Ph.D., the most popular use of the phrase "graduate school" actually refers to master's programs or professional schools, not Ph.D. programs. But Ph.D. programs have significantly different funding patterns from these other programs. Most business, law, medical and master's students fund their studies through loans and savings; in other words, past and future earnings are the most important financial resources of these students. Ph.D. students, on the other hand, are in programs significantly longer than students in other types of graduate schools (except medical school) and, though many take out large loans, earnings while in graduate school provide the bulk of the Ph.D. student's support.iii General HistoryThe first Ph.D. was granted in the United States in 1863, only 135 years ago. In 1871 throughout the United States, a nation almost 100 years old, only 44 Ph.D.s were granted. It was not until the 1890s, just about 100 years ago, that more than 1,000 Ph.D.s were granted in a year (Harmon). Not surprisingly, the number of Ph.D.s has increased dramatically over time. In almost every decade until the recent past, graduate enrollment has doubled or more than doubled (Syverson 1997a). Graduate enrollment is closely correlated with undergraduate enrollment; that is, roughly the same percentage of students, about 10%, who receive bachelor's degrees will go on to some form of graduate work in any given year (Harmon). As college degrees have become increasingly common, then so too have graduate degrees. PART TWO: DEMOGRAPHICS: WHO ARE GRADUATE STUDENTS?Peter D. Syverson, the Vice President for Research and Information Services of the Council of Graduate Schools, titles his 1997 report, "The New American Graduate Student: Challenge or Opportunity?" He makes the bold claim that "the students that [sic] are applying for and enrolling in our graduate programs are changing. The traditional view of graduate students as newly-minted bachelor's degree recipients engaged full-time in graduate study is contradicted by the data described below" (Syverson, 1997a). Today; more than half of Ph.D. students are married; 38% of doctoral students are women and about 25% have dependents other than a spouse. Approximately 50% of doctoral students are employed. Nearly half of doctoral students in the 1990s have education-related debt, either from undergraduate or graduate loans. The average doctoral student is carrying $16,800 in educational debt (Syverson, 1997a). In the following sections, I will compare this current demographic information to past demographic patterns. We will see that, indeed, significant changes have taken place since World War II which can help explain the rise in union activism. Gender Patterns of the Graduate PopulationFor most of this century, education has been particularly important for women. The gap between men's and women's wages has consistently been smaller at the higher end of the education spectrum than at the bottom: that is, a woman with only a high school diploma makes significantly less than her male counterpart, whereas a woman with a Ph.D. makes less than but much closer to the amount that her male counterpart makes (U.S. Census Bureau). Strangely, the history of women with graduate education has been all but forgotten. The Council of Graduate School's web page, for instance, reports the "trend" that women make up more than 50% of graduate enrollment in the United States (actually, 56% of master's degree students and 38% of Ph.D. students in 1995) (Syverson, 1997b). This is presented as a sort of 1990s feminist victory. However, as early as 1890, 18% of graduate students were female—that's nearly 1 in 5—and in 1928 women made up 40% (NSF). Graduate education reached an important high after World War II, when returning soldiers went back to school on the GI Bill. This is when the skewed sex ratio (on which the Council of Graduate Schools based its assumptions) really began; since the 1950s significantly more men than women attended graduate school, averaging about 7 out of 10 (NSF). The percentage of women has increased through the 1970s, 80s and 90s.iv Age Patterns of the Graduate PopulationIn the 1950s and 1960s, the average age of PhD recipients was 28 (NSF).v In 1974, the average Ph.D. recipient was 31. Today, the average Ph.D. student is 33 years of age (Syverson 1997a)vi. In fact, one-fifth of all graduate students are now over the age of 40 (Syverson 1997a). In other words, graduate students have aged during this century. Because graduate students with dependents "are more likely to be women and older than graduate students without dependents," the increasing numbers of female graduate students and the combined trends of admitting older graduates and the protracted time enrolled in a graduate program, has led to a graduate population with growing concerns about caring for dependents (Syverson 1997a) Spouses, partners and children are fixtures in the lives of more than half of graduate students. No wonder, then, that union campaigns, such as the one at University of Iowa, focus attention on dependent health care and child care demands. (Watt 247). Loan Patterns of the Graduate PopulationIn 1954, undergraduate loans were not a widespread phenomenon. Loans did not become an acceptable, mainstream method of financing higher education until late in this century. Indeed, the National Science Foundation (NSF), in its 1954 study of graduate student support, specifically limited its study of graduate school funding to teaching positions, research positions and fellowships. Loans were not considered important or influential in the financial lives of graduate students. Only in the 1980s did loans become a significant source of funding for graduate students. This explosion in borrowing happened simultaneously for undergraduates. Therefore, graduate students who borrow money for graduate eduction are even more financially burdened than their predessors because they are usually carrying undergraduate debt as well (Syverson 1997a). Timeline Patterns of Graduate School
Not only are graduate students coming to graduate school at later stages of life, they are also spending more time in graduate school. First let me present the ideal timeline given to anthropology students at the University of California, Santa Cruz:
Compare this to the 1954 National Science Foundation's version of the ideal timeline: In the ideal model, then, 50% of Ph.D.s were completed in three years, the remaining 50% in four years. In 1954, the ideal was 3.5 years; in 1998, the ideal at UC Santa Cruz was 5-6 years. In both cases, the ideal fell short of the national average by a year or two and the national average has been steadily increasing. In 1954, the national average was 4-5 years; in 1972, the national average was six years; today, getting a Ph.D. averages 7.2 years (NSF14-15; Cartter; Syverson, 1997a and Henderson). The extended duration of graduate school has many important implications, especially with respect to graduate school finances. Remember that Ph.D. funding is qualitatively different from professional or master's degree funding. Many of them use savings, family contributions or loans (future earnings) to pay for 2-3 years of graduate work. Cost-benefit analysis indicates that this is a smart move: future earnings will rise because of the degree and will, eventually, compensate for the loss of 2-3 years of salary and the tuition and living costs associated with graduate school (Princeton Review website). When the time period extends to seven years, however, this way of thinking becomes increasingly untenable.vii Employment Patterns of the Graduate PopulationNow we come to the crux of the argument as far as unions are concerned: the history of financial support. This is one of the areas in which the most dramatic changes have occurred. In 1954, approximately 11-25% of graduate students were employed while in graduate school, mostly as teaching assistants. In 1974, this number remained stable at 11-25% (Cartter, 1974). For 20 years, in other words, the percentage of employed graduate students remained the same. Then in the next twenty years the percentage more than doubled. More than 50% of Ph.D. students are now employed (Syverson 1997a). At many research schools, such as UC Santa Cruz, the number is even higher, around 80% or more (60-65% as teaching assistants; 20-30% as research assistants)(Syverson 1997a and ASE/UAW). Research assistantships and teaching appointments are not uniformly distributed across the disciplines, however, and never have been. In the section that follows, I will show some of the history of the funding pattern disparity between disciplines. Then I will examine who among graduate students tends to reap the benefits and who is negatively impacted by this disciplinary hierarchy. The 1954 NSF study points to a dramatic difference in the distribution of "support" between the natural sciences and all other Ph.D. programs. "About 6 out of every 10 resident graduate students of the natural sciences received support—twice as many as in any other major field" (NSF 19). In the social sciences, about 2-3 out of ten students received support. Support was defined as a) a teaching position, b) a research position, or c) a fellowship, which required no work in exchange. Fellowships, the most desirable form of support, made up a rather small percentage of the entire support studied. Research positions were the next most desirable form of support because they paid significantly more than teaching positions. These three forms of support were not uniformly distributed, however. As an example, I will compare the fields of English and chemical engineering in 1954. In English departments across the United States approximately 30% of the students received some money from their institution: 23% were teaching assistants; 1% were research assistants (making more money than the TAs); and 6% were on a fellowship. In chemical engineering, by contrast, approximately 43% received some form of money: 12% as teachers, 17% as researchers and 14% as fellows (NSF 1954). Employment Patterns of the Graduate Population: Race and GenderAnalyzed in racial and gender terms, such disciplinary disparity reveals a strong division between the "haves" and the "have-nots" of today's graduate students. White men (and sometimes Asian-American students) in the 1990s tend to be enrolled in disciplines whose graduates receive proportionately more money.viii This money comes in the form of fellowships and research assistantships as opposed to teaching assistantships. By contrast, white women and Native American, African-American and Chicano/a men and women tend to be enrolled in disciplines whose graduates receive less money overall, and receive money mostly in exchange for teaching. In 1996, 70% of graduate students in the UC were white and 58% were men. The female and non-white students were concentrated in the arts, humanities and social sciences. Nearly 60% of women were enrolled in these disciplines as were 65% of Native American, African-American and Chicano/a graduates ("Composition," 8). The University of California spent an average of $10,782 per graduate student per year on engineering and computer science research positions. By contrast, average spending for humanities RAs was a paltry $868 and for social science RAs $2,065. To make up the difference, humanties and social science graduates taught in far greater numbers, receiving per graduate an average $8,841 and $6,605 respectively.ix Engineering and computer science graduates worked far fewer hours as TAs, receiving on average only $2,231 for teaching ("Annual Report").x In the end, social science and humanities graduate students received an average of $1,000 less per year than UC's estimated budget for graduate students. Natural science graduates received $3,000 more than the estimated budget. PART THREE: THE ROLE OF TEACHING ASSISTANTS IN THE ACADEMYUniversities Increase Dependence on Teaching AssistantsDespite these significant disciplinary variations, teaching while in graduate school has increased dramatically in all disciplines in the past twenty years. In 1954 "teaching assistantships were the preponderant mode of support not only in education and the humanities, but also in the physical sciences, psychology, and the social sciences.... [In] the social sciences, the proportion of resident students with teaching assistantships was closer to the national average of 11%" (NSF 22). Through the 1960s and early 1970s, this national average remained at 11% (Cartter 108). Twenty years later, the average percentage of Ph.D. students engaged in teaching has risen from 11% to 50% or more.xi At UC Santa Cruz, for instance, there are approximately 850 Ph.D. students. In one academic year, more than 556 of them work as teaching assistants, or close to 65%. In addition, TAs provide about 60% of the contact teaching hours with undergraduates in the UC system (ASE/UAW). This jump from 11% to 50% or more has created the critical mass necessary for a collective identity and movement. Though graduate teachers were certainly teachers in earlier years, the sheer number of TAs today has contributed to the success of the union movement. Post-Graduate School EmploymentThe employment patterns of those who have completed the Ph.D. contrasts sharply with the above statistics. In fact, the number of Ph.D.s employed in academia seems to have moved in inverse proportion with the teaching they did while in graduate school. That is, when the percentage of graduate students who were hired as TAs was low, the percentage of graduate students who were eventually hired as professors was high. In 1958, 70% of Ph.D.s found work in teaching or academic research. In 1970, 74% found work in teaching or research (Cartter 190). In the 1990s, by contrast, while the TA rate hovers around 50-60%, 33% or less of Ph.D.s—and even lower rates exist in the humanities—find secure jobs in academia (Young 184; Paymar 42). The fundamental assertion of university administrations who fight academic student employee unionization, including most notably the fifteen-year stance of the University of California, is that teaching assistants are receiving job training as apprentices. This assertion just doesn't make sense in 1999.xii The Importance of TAs to the UniversityWhy has the percentage of graduate students employed as teachers gone up so dramatically? Teaching assistants have been integral to the university system throughout this century. Back in 1954, when only 11% of graduate students were TAs, their importance to the university was declared by the NSF. Their study states that while [being a TA] may necessitate part-time study and thereby delay the acquisition of a degree, the benefits derived by both student and institution have made assistantships a vital element in this country's system of higher education. Insofar as experience in instruction and educational methods is gained, assistantships help prepare the graduate student for a future academic career. At the same time, assistants relieve the institution's staff of what would otherwise be an overwhelming burden. Without graduate teaching assistants, some of the nation's greatest schools could hardly continue to function as they do today... (NSF 22). This same sentiment was expressed by Allan M. Cartter in his 1976 book, Ph.D.'s and the Academic Labor Market. Cartter writes that TAs are "often of high caliber as teachers. Indeed, most of them will be full-time faculty in another year or two" (Cartter 108). These statements were made when only 11% of graduate students provided labor as TAs. But let us return to the basic question: why do universities use teaching assistants? After all, expensive faculty are hired, ostensibly, to teach undergraduates. Why not let them do this job entirely? There are at least two possible answers to this question and they are by no means mutually exclusive. The first answer, which the UC administration stresses in its court cases against the unions, is that teaching assistanships are providing job training. According to this argument, Ph.D. programs produce college professors and being a TA prepares graduate students for their future careers. The second answer, and one which Cartter states explicitly in 1976, is that graduate student teachers are simply cheaper and easier to hire and fire (more "flexible") than full-time faculty. Cartter notes that they represent a kind of captive reserve labor supply—they can easily be laid off or displaced depending on the vagaries of undergraduate enrollment, and their salary rates are set more by fellowship standards than by going wage rates. Most instructors in this group have not yet entered the market in the sense of seeking and weighing alternative offers. (Cartter, 1974:108. Emphasis mine) Graduate student labor is cheap relative to full-time faculty labor. Replacing TAs at Yale, for instance, "with junior faculty members would cost the university $5.5 million per year" (Entin, 1997:27). Perhaps a more important factor is graduate student flexibility. Cartter explains that the nature of academic work parallels that of "investment goods in the economy as a whole" (Cartter 1974:2). That is, if the ratio of students to faculty remains constant at, say, 15 to 1 and total enrollment moves from 5,000,000 to 5,150,000, one can expect that about 10,000 new teachers will be required to handle the additional students. However, if instead of a 3 percent increase in enrollment there is a 6 percent increment, the total number of new teachers required to handle new enrollments would rise by 100 percent. Therefore, like the case of investment goods, small changes in the demand for final products (the education of students) produces an exaggerated change in the demand for investment inputs (college teachers). Fairly significant swings in the demand for new academic personnel must be expected in the academic labor market (Cartter 2). Following this economic logic, Cartter reasons that there are at least four methods academic institutions can employ to handle changing demands. If there is a downswing in enrollment, administrators can promote early retirement; if there is an upswing in enrollment, administrators can change hiring standards to hire M.A. level teachers. To prepare for either up or down- swings in enrollment, the administration can adjust student-faculty ratios or it can have a "greater or lesser reliance upon adjunct and junior instructors" (Cartter 2). However, this economic model was equally true before the dramatic increase in university reliance on teaching assistants of the last two decades and remains true in elementary and high school education. What appears to have shifted recently, then, is not the reasons for hiring teaching assistants (their relative cheapness and flexibility), but the acceptability of this practice. PART FOUR: CHANGING GRADUATE EXPERIENCE LEADS TO UNIONS IN THE ACADEMYSo let us return to the main argument, then, of labor unions within the academy. My foray into history has convinced me that it is no accident that the birth of graduate student employee unions has occurred in the last twenty years. In this period, the number of TAs as a percentage of graduate students has risen dramatically, and, with it, the percentage of the college curriculum taught by graduate students and not professors. Whereas in 1974, only 1 in 10 graduate students was a TA, today the figure is 1 in 2. It is not that a TA was any less an employee in 1954 or 1974, but there was probably not sufficient critical mass to begin a union movement. Perhaps more importantly, the average graduate student in 1954 or 1974 received a job within academia. Therefore, the argument which the UC administration uses in 1999—that TAs are receiving job training—was three or four times more applicable in the past than it is today, when only 33% of Ph.D.s will receive academic jobs. Strangely, in the past, when more than 70% of Ph.D.s did receive academic jobs, the university did not seem to think it so important to provide all graduate students with teaching experience. In fact, the situation is almost the reverse of 1954. Ted, an English Ph.D. student at the University of Iowa, says that "people stay in graduate school because they think, 'This is my last chance to teach at a university'" (Interview). In addition, the fact that the average length of time to complete a Ph.D. is 7.2 years and that the average age of graduate students is 33 changes the way that graduates look at graduate school. A 23-year old looking at four years of graduate school will think of these years one way. A 33-year old with a husband, four-year old son, and $16,800 of debt will think of seven years out of her life quite differently. Especially when the employment percentage is factored in, these seven years have to start counting for graduate students as more than just job training. As Joseph Entin, a graduate student at Yale, says, "Why pay dues to a club that might never admit you?"xiii Union dues—and union contracts—are the alternative of choice for hard-working and dedicated teachers who also happen to be graduate students. The University of California UAW union victories point the way to a new kind of university. Notes
Sources Cited
|