Economically, it would seem that faculty have much to be happy about. The academic year 2001–02 was the fifth consecutive year in which the value of the average faculty salary rose—and one in which academics saw the largest single-year jump in their real (inflation-adjusted) salaries since the mid-1980s.1 The increase in nominal, or actual, salaries between 2000–01 and 2001–02 averaged 3.8 percent, while the rate of inflation at the consumer level between December 2000 and December 2001 was 1.6 percent, meaning that the average professor had 2.2 percent more purchasing power this year than last.
Is this the beginning of a new, rosier future for faculty members? Unfortunately, it probably is not. The relatively large increase in 2001–02 results from the unusual timing of inflation in the United States during the past two years. A rapid increase in consumer prices during the first half of 2000 meant that the inflation-adjusted increase in faculty salaries between 1999–2000 and 2000–01 was small—only 0.1 percent. During the second half of 2001, however, the level of consumer prices actually dropped slightly, so that price inflation over the whole year averaged out quite low.
It is possible that consumer price inflation during 2002 will average as low as it did in 2001, but two considerations make it unlikely that faculty salaries will rise at last year’s rate of 3.8 percent. First, institutional administrators now believe that inflation will be lower than last year, and they will therefore probably set nominal salary increases lower than last year’s so as not to overcompensate faculty in relation to the expected rate of inflation. Second, most institutional budgets for 2001–02 were set by June 2001, before the recession really hit state and local tax revenues. Budgets for 2002–03, however, are being set right after a recession and at a time when most states and localities have seen constant or even declining tax revenues.
Table 1 (.pdf) shows salary increases by rank for the past thirty years. From 1971–72 through 1985–86, the increases are calculated over two-year periods; thereafter, they are presented annually. The percentages shown under the heading Nominal Terms reflect increases in actual average salaries; those under the heading Real Terms are adjusted according to the Consumer Price Index (CPI) to account for inflation at the consumer level. The All Faculty section of the table shows that salary increases in 2001–02 were highest among assistant professors. Given the relative tightness of academic labor markets, that makes sense. The market for assistant professors is affected most strongly by alternatives outside academe, which were relatively plentiful until recently. To meet the competition for entry-level talent without reducing quality, colleges and universities had to raise salaries. The market for junior faculty, where most of the hiring (and firing and quitting) occurs, is the bellwether of the academic labor market generally.
The Continuing Faculty section of the table reports percentage salary increases only for those faculty members who remained at the same institution over the two academic years shown. As in all the previous years for which we have data, the rate of the salary increase for continuing faculty in 2001–02 is above that for the average faculty member. In fact, in the All Ranks category, this year’s 1.2 percent difference between the Continuing Faculty increase (5.0 percent) and the All Faculty increase (3.8 percent) is almost identical to the average difference (1.17 percent) over the entire thirty-year period for which the table presents data.
Why is the figure for Continuing Faculty always higher than that for All Faculty? Each year, the average campus will see older faculty members retire, younger ones hired, and the rest age by one year. The generally higher-paid older faculty who retire are replaced by new, younger faculty, most of whom receive lower salaries. The salary figures for All Faculty are affected by this turnover, which lowers the average salary on campus. The salary increases received by continuing faculty, both those who remain within rank and those who are promoted to a higher rank, are also included in the All Faculty figures. But the figures for Continuing Faculty include only the increases received by faculty remaining on campus; they exclude the generally lower salaries of first-year faculty.
Which figure should you use to assess the economic status of the profession? If you want to examine the well-being of the profession as a whole, the All Faculty figure is relevant: it best reflects what happens to the salary of the "typical" faculty member—the average person on the average campus. The Continuing Faculty figure is, however, relevant if you want to analyze how faculty members on your campus who are aging one year each are faring compared with the other faculty members nationally who are moving up the ranks (or at least garnering more experience within rank on their campuses).
A good way to see the difference between the All Faculty and the Continuing Faculty measures is to note that among All Faculty, no one academic rank consistently has a higher percentage increase than the others. In some years, such as the current one, assistant professors have the highest rate of increase; in other years, such as 1999–2000, full professors had it. In yet others, such as 1996–97, associate professors received it. For continuing faculty, however, the rates of increase are generally highest among assistant professors, lowest among full professors. As we move up the ranks on our campuses, our rates of salary growth tend to slow, a phenomenon that makes us no different from employees in every other occupation in the public and the private sectors of most developed economies.
The data in table 1 are hard to digest—forming a coherent picture of what has been happening in academe over the past thirty years from all the annual salary increases during that time is difficult. Figure 1 (.pdf) helps fill out the picture by showing changes in the inflation-adjusted salary of the average full-time faculty member from 1971–72 through 2001–02. The figure uses 1971–72 as the starting point (arbitrarily set at 100), and the rectangle denotes the average salary adjusted for inflation in consumer prices, as measured by the CPI.2 The line makes clear what should by now be a well-known fact: the real pay of the average academic fell between 1971-72 and 1981-82. The substantial drop in faculty salaries was produced by an oil shock that generated a severe recession and by the end of the baby-boom bulge of college enrollments. Since 1981-82, real salaries, adjusted for inflation using the CPI, have risen nearly steadily. The rise over the past five years in particular has brought the average level of real salaries very slightly above that for 1971-72 for the first time in thirty years.
This report has always presented data showing actual salaries deflated by the CPI to obtain a measure of inflation-adjusted salaries. The CPI is only one of several measures of inflation that can be used to make this adjustment. For a variety of reasons, some abstruse, others straightforward—for example, a failure to account properly for the introduction of new products or for improvements in the quality of existing products—many of these indexes overestimate the rate of increase in average prices.3 In fact, evidence suggests that the CPI has often overstated inflation by as much as 1 percent a year. Other measures of consumer prices also have problems, but they avoid at least some of those that were inherent in the CPI until the federal government recently revamped it.
The line in figure 1 denoted by the triangle presents the salary of the average academic adjusted for inflation using the deflator for Personal Consumption Expenditures (PCE) from the National Income and Product Accounts, the source of information on the nation’s output and income produced by the U.S. Department of Commerce’s Bureau of Economic Analysis. Like CPI-adjusted salaries, PCE-deflated pay fell during the 1970s. There is little doubt that the average academic was worse off by the end of the 1970s than at the start. Indeed, even when the PCE deflator is used, the real salary of the average academic did not return to its 1971–72 level until 1989–90. Like CPI-adjusted salaries, PCE-deflated salaries rose during the 1990s—but at a faster rate than real salaries calculated according to the CPI. Measurements based on the PCE deflator indicate that the average academic earns 13 percent more today than his or her counterpart earned in inflation-adjusted dollars in 1971–72.
This better measure suggests some improvement in the real earnings of academics over the past thirty years; but the average rate of increase in earning power, even by this measure, has been paltry: only 0.4 percent a year. So the only possible conclusion—the best face one can put on the situation—is that academic salaries have improved very, very slightly over the past three decades. Moreover, these adjustments affect only the interpretation of changes in the level of faculty pay. They cannot obscure the fact that faculty salaries have fallen relative to those in other professions, a topic discussed later in this report.
Rank, Institutional Type, and Region
The data in table 1 allow us to examine changes in pay levels across academic ranks over the past three decades. One study noted a remarkable constancy in relative pay between full and assistant professors, full and associate professors, and associate and assistant professors from the mid-1960s all the way through the mid-1980s.4 Figure 2 (.pdf) updates the comparison through the current academic year. The constancy in relative pay across ranks persisted through 1994–95: the pay ratio of full to assistant professors hovered around 1.62, while that of full professors to associate professors varied in a narrow range around 1.34.
Since 1994–95, salary differences in academe have departed from these historical constants. As figure 2 shows, the pay of full professors relative to both associate and assistant professors has risen, while pay differences between the two lower ranks have remained essentially unchanged. The relative pay of full professors rose to about 1.65 times that of assistant professors, and to 1.38 times that of associates. In the tight labor market of the 1990s, it was the pay of the more senior people, those who would seem less likely to leave academe, that inexplicably increased the most.
Other AAUP salary reports have noted the widening pay disadvantage among faculty at public colleges and universities compared with those at private institutions. The causes of the relative decline in public-sector pay in academe are unclear. The most likely explanation is the increased unwillingness of taxpayers, and through them legislatures, to spend more money on what was once a well-regarded task of state and local governments.5 The evidence on this relative decline in prior AAUP reports and elsewhere is from the mid-1970s through the mid-1990s. But what has been happening since then: has the relative decline continued, or has the trend reversed?
Table 2 (.pdf) presents percentage increases in average salaries since 1981–82, since 1991–92, and since 1996–97. The data from 1981–82 and 1991–92 show that salaries rose more slowly in public than in private higher education during the 1980s and the 1990s.6 During the 1990s, however, the relative decline in public-sector pay was concentrated entirely in the first half of the decade: between 1996–97 and 2001–02, pay in both sectors rose at almost identical rates. The downward trend in relative pay did not reverse, but it did cease. These results are heartening for faculty in the public sector. The question remains, however, whether public higher education has simply had a respite resulting from the flush state budgets of the late 1990s, or whether its increasing relative impoverishment has finally stopped. Experience over the next two years, with the expected tightening of state budgets, will answer that question.
Table 2 also analyzes increases in average salaries by institutional category, ranging from doctoral-level (Category I) institutions to colleges and universities without ranks (Category IV).7 If we set aside institutions without ranks, the calculations show that pay gaps have steadily widened across institutional types. Pay at doctoral-level universities, already higher than that in other categories in 1981–82, rose more rapidly in percentage terms than salaries in the other categories in each of the periods examined in the table. The rate of increase at doctoral institutions was followed closely by that at general baccalaureate (Category IIB) institutions, while comprehensive (Category IIA) institutions, two-year colleges with ranks (Category III), and institutions without ranks saw smaller increases. The apparently anomalous result for general baccalaureate schools arises because many of them are private liberal-arts colleges that, like private institutions generally, saw relatively high pay increases through the mid-1990s.
Over these decades, different regions of the United States have experienced different economic shocks and, with them, different ups and downs in the government revenues that support public higher education as well as in the resources that support private institutions. These varying fortunes, which have affected regional increases in academic salaries, are reflected in the statistics presented in Figure 3 (.pdf). For each of the nine official subregions of the country—New England, Middle Atlantic, East North Central, West North Central, East South Central, West South Central, South Atlantic, Mountain, and Pacific—the figure presents five-year percentage changes in average salaries from 1986–87 to 2001–02.
Institutions in the Pacific suffered during the economic shocks of the late 1980s and early 1990s, and the pay increases of the past five years have not restored that region’s premier position. Conversely, New England saw the second-largest increase among regions in average salaries in the late 1980s and the highest increase in the past five years. Some regions do well in some periods, others fare better in other periods, but the effects of these fluctuations accumulate systematically. The result (not shown in the figure) is that the New England, Middle Atlantic, and South Atlantic regions have seen the biggest percentage increases in average salaries over the past fifteen years, while the East South Central and Pacific regions have seen the smallest increases. The Pacific region, which had the highest average salaries among the regions in 1986–87, now ranks second highest, while the East South Central region, lowest in 1986–87, has fallen still further behind.
Academia, Other Professions, and Gender
The AAUP salary survey covers academic institutions. To understand how faculty are faring compared with professionals outside academia, we need an additional source of data. Fortunately, the Current Population Survey (CPS), the same monthly survey that provides information on the nation’s unemployment rate, has obtained consistent information on workers’ occupations and usual earnings since 1979.8 In the calculations for this report, I included only those individuals in the CPS who worked at least thirty-five hours a week in their occupations and who had at least a master’s degree. Figure 4 (.pdf) shows the ratio of earnings in each of four groups of professionals outside academe—health professionals, scientists, engineers, and attorneys—relative to those of college and university teachers for 1980, 1985, 1990, and 1994–2000. Except for the early 1990s, when scientists’ average earnings equaled those of faculty, the earnings of all the other groups have consistently exceeded those of faculty. Also noteworthy is the generally rising relative pay in the four other occupations compared with that in academe over the past two or three years. There is little doubt that college and university faculty lost ground to other professionals beginning in the late 1990s. Compared with our relative rewards in the mid-1990s and even 1980, we are today less well rewarded than many other comparably educated professionals.
Within this relative poverty, have academic institutions at least made progress compared with employers of other professionals in how they pay their female members and in the numbers of women they hire? The same CPS data allow us to answer this question. Because the samples are fairly small, I combined the years 1979 and 1980, 1984 and 1985, and so on through 1999 and 2000. Figure 5 (.pdf) shows the ratio of male to female faculty earnings in academe and in the same four professions that were compared with academe in figure 4. The picture is one of no huge differences in male-female pay among the professions. Neither does the figure indicate any obvious trend in gender pay differences, either in academe or in most of the other professions. In only one profession—engineering, where women are scarcest among the five professions examined—is there any clear trend toward declining male-female pay differences. Progress in enhancing female pay relative to that of males has been as lacking in our profession as in most others.
Figure 6 (.pdf) looks at whether women are becoming more visible in academe compared with other professions. Over the past twenty years, the share of women among college and university teachers, shown in the figure by the rectangle, has risen from 25 to 37 percent. This is a substantial and welcome increase. At the same time, however, the presence of women among all professional and technical workers, shown by the circle in the figure, has also increased greatly. Thus the differences between the growing percentages of women in academe and among all professions have been slight.
Between 1960–61 and the mid-1980s, the share of fringe benefits in academic compensation roughly tripled, paralleling a national explosion in compensation spending for retirement, health insurance, and social security.9 After a brief pause, fringe benefits as a share of academic pay rose again through the early 1990s, as the first two columns of table 3 (.pdf) show. Since the early 1990s, however, there has been essentially no change in the share of faculty compensation accounted for by fringe benefits. While spending on medical insurance has risen slightly as a percentage of salary, spending on retirement plans has fallen slightly.
Economists have attributed the national rise in fringe benefits between the 1950s and the early 1990s to a combination of several factors: the tax treatment of the benefits at a time of rising tax rates, economies of scale in administering benefit plans, and higher real incomes that made workers want more of their potential income in the form of benefits rather than earnings. Tax rates have not fallen in the past decade, it is still easy to administer benefit plans for large groups, and table 1 and figure 1 show that real earnings in academe have risen. Yet for some unknown reason, these forces are no longer combining in a way that continues to raise the share of fringe benefits in faculty compensation.
Having produced this report from 1992 through 1996, I was asked to resume the assignment for this academic year. The AAUP’s director of research, Ernst Benjamin, who had overseen the salary survey for many years, had recently retired, and the Association was in the process of searching for a new research director. In the interim, neither this temporary assignment nor the survey itself could have been completed successfully without the hard work and dedication of the AAUP’s research associate, Galina Lewis, who supervised the collection and input of data and produced the survey report tables and appendices. Ernst Benjamin, who now serves as senior consultant to the Association, ensured that I received the survey report tables with the alacrity that allowed me to generate the report to meet Academe’s publication deadlines, and he also suggested one of the topics on which this report concentrates.
I also wish to thank the members of the Association’s Committee on the Economic Status of the Profession, several of whom offered comments on a preliminary draft of this report that altered my interpretations of the findings. The committee members are Linda A. Bell (Economics), Haverford College; Robert C. Comeau (English), Union County College; Anne Harrison (Business), Columbia University; George Lang (Mathematics and Computer Science), Fairfield University; Steven London (Political Science), Brooklyn College, City University of New York; James May (Communication Science and Technology), California State University–Monterey Bay; and Craig Swan (Economics), University of Minnesota, consultant.
DANIEL S. HAMERMESH
Centennial Professor of Economics
University of Texas at Austin,
And Chair Committee on the Economic Status of the Profession
1. Most of the information in this report is based on the AAUP survey of higher education institutions in the United States. In 2001–02, 1,433 institutions (representing 1,700 campuses) are represented in the survey. Data from these institutions are included in the basic results in table 1 and many of the other tables in this report. AAUP staff compiled the data on which the tables in this report and the appendices that follow are based. Because of changes over the past two years in the way the National Center for Education Statistics (NCES) collects faculty salary data, we have not been able to draw on responses prepared for the NCES to the extent that we did prior to 2000–01. Consequently, there are fewer institutions represented in the surveys for 2001–02 and 2000–01 than in those for previous years. This decline, especially notable for the representation of community colleges, may modestly affect the reliability of some items. Survey Report Tables 14a and 14b for this year may be compared to the same numbered tables for previous years to assess the extent of diminished participation by institutional category and affiliation. Back to text
2. This calculation reflects a compounding of the data in the penultimate column of the All Faculty section of table 1. Back to text
3. Michael Boskin et al., "The CPI Commission: Findings and Recommendations," American Economic Review 87 (May 1997): 78–83, summarizes the report of the commission. Back to text
4. Daniel Hamermesh, "Salaries: Disciplinary Differences and Rank Injustices," Academe (May–June 1988). Back to text
5. The January 2002 issue of the Postsecondary Education Opportunity Newsletter reports that state tax spending for higher education was $9.97 per $1,000 of personal income in fiscal 1982, $8.24 per $1,000 in fiscal 1992, and only $7.67 per $1,000 in fiscal 2002. Back to text
6. Cynthia Zoghi, "Why Have Public University Professors Done So Badly?" Economics of Education Review 21 (2002), shows that the same trend also existed over this period adjusted for the degree of competitiveness of the institutions. Back to text
7. A definition of institutional categories appears in the Explanation of Statistical Data. Back to text
8. The Current Population Survey is produced by the U.S. Bureau of Labor Statistics. Back to text
9. See Stephen Woodbury and Daniel Hamermesh, "Taxes, Fringe Benefits, and Faculty," Review of Economics and Statistics 74 (1992): 287–96. Back to text