Executive Summary of the Economic Status of the Faculty 2010-2011 Report	Print Issue
February 28, 2012, Volume 58, No. 24

The full report of the Economic Status of the Faculty 2010-2011 Report is available as a PDF only.

Introduction
This Executive Summary is meant to cover the most salient portions of the full Economic Status of the Faculty 2010-2011 Report. This report addresses salary increases in 2010-2011, not in 2011-2012, which will be the subject of the report next year. 

The Summary concludes with the Committee’s Recommendations and Questions for the Administration for 2010-2011. The Committee encourages readers to access and review the complete report.

FY 2010-2011 followed the unusual events of financial crisis and depressed value of, and returns on, the University endowment in the prior year. For all ranks combined, the mean FY 2011 percentage salary increase was just slightly larger than the percentage change in the U.S. City Average CPI and one point higher than the Philadelphia CPI. However, the all-ranks median figure is considerably below the U.S. City Average CPI growth and below the Philadelphia CPI as well while being exactly in line with the university salary guidelines.

Comparisons with Peer Universities Using Data from the AAU Data Exchange
The best currently available salary data from other institutions of higher education are provided by the American Association of Universities (AAU) Data Exchange. The AAU is comprised of 60 public and private research universities in the United States and two in Canada. The AAU includes several Ivy League institutions (e.g., Penn, Brown, Harvard, Princeton, Cornell, and Yale), other private universities (e.g., Brandeis, Rice, Emory, Vanderbilt), public flagship universities (e.g., Berkeley, UCLA, the Universities of Michigan, Virginia, and Wisconsin), and other public universities (e.g., Michigan State, University of California Davis, and University of California Irvine). Please refer to the AAU website for a complete list of member institutions: www.aau.edu

Salary Comparisons: Penn’s Competitive Standing
The most meaningful comparisons of mean faculty salaries at Penn with those at other universities in the AAU Data Exchange are broken out by academic field and rank. 

For most of the 16 schools and areas, Penn’s mean faculty salaries for all ranks in 2010-2011 rank in the upper fifth of the AAU institutions. The exceptions are the Natural Sciences in SAS, the Humanities in SAS at the Assistant and Associate Professor ranks, the School of Veterinary Medicine for all faculty ranks, the School of Social Policy and Practice in the Assistant and Full Professor ranks, and SEAS at the Full Professor rank. 

Some of these lower salary rankings, in SAS and SEAS for example, have remained fairly stable over time, pointing to a chronic issue. Such is the case for the School of Veterinary Medicine. Assistant Professors are not far off from the top third but rank slightly lower this year than last, while Associate Professors rank well in the bottom third, showing as they did last year a marked decline.

For the rest of the schools and ranks, the general pattern of relatively high overall position vis-à-vis the AAU comparison institutions might be seen as acceptable if not for trends over time. Examination of Fall 2006 comparisons with those for Fall 2010 reveals a worrisome trend toward comparative stagnation or frank decline in rank order within several schools and areas. This pattern has been addressed in SCESF Reports for the past two years. It remains unchanged. While Penn has made some gains in AAU rank comparisons, it continues to fall behind in many more.

Comparisons with Peer Universities Using AAUP Survey Data
The table below presents a comparison of the mean salaries of all full professors at Penn with those at a small select group of research universities based on data obtained by the Penn administration, collected annually by the American Association of University Professors (AAUP), and published in the Chronicle of Higher Education. 

The data in this table show that, during the past five-year period, mean salaries for Full Professors at Penn became more competitive with some few institutions in the comparison set (four in total, though only two by margins much greater than rounding error) but became less competitive with the overwhelming bulk of the panel. Again, we give a more detailed analysis in the longer version of our report.

Variability in Average Salary Levels by Rank
Data on mean and median faculty salaries by rank for all schools combined are shown in the following table for each of the past five years: 2006-2007 through 2010-2011. The second-to-last column gives raw ratios of these values relative to the values for Assistant Professors. These ratios imply that, in FY 2011, mean salaries were 69% higher for Full Professors than for Assistant Professors and 7% higher for Associate Professors than for Assistant Professors. Median salaries were 85% higher for full than for Assistant Professors, and 15% higher for Associate than for Assistant Professors. The longer version of our report provides a more extensive discussion of these data.

Trends in Variability Over Time
The measure of variability of median salaries across schools/areas of continuing faculty members selected here is the interquartile range (IQR) (i.e., the 75th percentile salary in the distribution less the 25th percentile salary). However, the IQR can be expected to be larger when the general salary level is relatively high (such as for Full Professors) than when the general salary level is much lower (such as for Assistant Professors). To compensate for such differences in the general level of salaries, we have divided the IQR by the median of the distribution (i.e., the 50th percentile salary: Q2), thereby computing a ratio of the IQR to the median (as reported in the next to last column of Table 10 labeled “IQR to Median” in the full report). This ratio provides an index of the amount of variability in relation to the general level of the salary distributions, and has utility when comparing variability across ranks and trends over time.

The most striking feature of the analysis of variability is the rise over the past five years of the IQR to median ratio, particularly for Full Professors, suggesting that variability in payment levels is growing (from 0.42 to 0.46.) For Full Professors, in 2006-2007 the gap between the 25th and 75th percentile was 59,900; in 201l that gap had grown to 74,131 (a 23.8% rise.) The median salary had only risen 12.8% over the same period. For Associate Professors the variability change was smaller and not consistent but also showed an increasing gap. (The IQR grew 16.3% while the median grew on 9.3%.) The Assistant Professor variability was much less stable in trend than for the other ranks and thus the 2010-2011 increase in variability is difficult to interpret. However there does appear to be evidence for increasing spread in faculty salaries and warrants further exploration.

Variability by Gender
The first table provides the percentage increases in salaries for faculty continuing in rank by rank and gender for the first, second, and third quartiles for FY 2011. The figures are generally very close and in some cases identical. Only in one case is there a meaningful difference, for Assistant Professors at the median (3.0% for men and 2.5% for women).

The second table reports unweighted and weighted observed mean and median salaries for men and women continuing in rank by rank. The weighted estimates account for difference in gender distributions in schools and areas of the University and thus may be most informative. The weighted analyses show that male Full Professors are consistently better paid than female Full Professors (e.g. 4.8% higher mean in 2010-2011), however this advantage appears to be declining (from 6.9% in 2006-2007). The male advantage for Associate Professors is less consistent. It is always present in the means but mostly reversed for median comparisons, and also appears to be declining. For Assistant Professors, the differences between the mean salaries by gender are small. However, there appears to be a small increase in the gap by 2009-2011 compared to previous years. 

The Committee recognizes that neither the gap closing nor gap opening trends are definitive without additional information about years in rank and departmental-level weighting and other relevant information.  Importantly, the trend for Full Professors shows some promise for continued improvements. Thus the Committee looks forward to a subsequent and more elaborate analysis of gender equity among faculty.

The complete version of the report gives extensive discussion of the current economic status of the faculty and conditions of concern going forward.

SCESF Recommendations and Questions for the Administration for 2010-2011

In accordance with Faculty Senate policy, following are recommendations and questions for the administration that arose in the SCESF discussions, including some updates on the status of recommendations made in previous SCESF reports.

1. Salary Competitiveness
To provide high-quality instruction, research, and service, the University must maintain and attain faculty salaries at levels that are highly competitive with salaries provided by peer universities, while simultaneously sustaining other components of university operations.

SCESF Recommendations:
a) Mean salaries at Penn have fallen behind in the comparison with AAUDE data in a number of schools and areas (e.g., compare first and last columns in Table 4 on either the traditional or the more robust measure discussed above). The SCESF recommends that priority be placed on increasing mean salaries to competitive levels for the faculty groups that have fallen behind. 

The President and Provost remain committed to enhancing Penn’s ability to offer highly competitive faculty salaries, while recognizing that some of our peers enjoy greater financial resources than Penn. The Provost’s Office agrees to explore apparent changes in the competitive standing of fields identified by the SCESF report and to explore with deans actions that may be justified and financially feasible.

b) SCESF notes that there is room for improvement for faculty salaries in many of the rank by school and area comparisons (Table 4). Moreover, the gaps in mean salaries between Full Professors at Penn and Full Professors at some peer institutions have increased over time see Table 5. SCESF questions whether the University can retain and attract the highest-quality faculty members unless faculty salaries improve markedly in relation to peer institutions. Our competitors seem to perceive an active market for university faculty. Given the quality and visibility of the Penn faculty overall, current faculty may easily be swayed with this market perspective to consider other options. Prospective faculty too may be drawn to competitors. Continued use of aggressive responses to individual retention problems are not a viable response to this persistent pattern of declining external competitiveness, especially when coupled with internal inequity. This larger problem mandates a thorough University-wide response.

We note that the vast majority of Penn’s salaries for all ranks and schools remain within the top ten institutions of their cohort (32 of 39 comparisons). Comparisons over time for particular ranks and schools on the basis of AAUDE data are problematic because of changing numbers and identities of participating schools and shifting composition of the disciplinary categories. In addition, the reported mean salaries do not take into account adjustments for relative costs of living, which are significantly higher in many of the cities where our peer universities are located. We will continue to closely monitor the competitive standing of Penn salaries.

c) The SCESF places priority on regaining Penn’s competitive level in the fields where it has been declining or lost. Nevertheless, the SCESF recommends that priority and due consideration also be given to recognizing and rewarding distinguished performance with salary increases for those faculty who choose not to seek, or use, attractive offers of external appointment to negotiate salary increases. This approach is a matter of equity, morale, and investment in a secure future for the University. As stated in the SCESF Report last year, the SCESF recognizes that these are decisions taken at the Dean and Department Chair levels but observes that decision-makers at those levels are often keenly aware of budget constraint issues. The SCESF recognizes that the Guidelines published annually are explicit on this matter but feels that explicit additional guidance from the Provost downwards would be very helpful in this matter.

Salary increases are awarded within Schools on the basis of merit, and “distinguished performance” constitutes merit. The Provost will continue to discuss with School Deans issues posed by market competition, retention, and merit within school faculties. School Deans are aware of equity issues and regularly respond to them. Budget constraint issues operate at all levels of the University.

2. Salary Equity
Inequity among individual faculty salaries by rank within departments and schools organized as single departments must be identified and eliminated.

SCESF Recommendations:
a) The SCESF continues to recommend that the Provost and Deans give consideration to decreasing the number of instances in which faculty members who have performed at least at a satisfactory level are awarded salary increases that are below the annual growth in the Philadelphia CPI.  With this recommendation, we realize that the feasibility of awarding increases to faculty members with satisfactory performance at least as great as growth in the CPI depends on the difference between funds available for salary increases and the CPI growth percentage. We also recognize that there may be periods of financial stringency as in the FY2009-2010.   Nonetheless, we continue to fear that extended declines in real wages and loss of salary against CPI will have marked adverse effects on faculty morale and University function. 

Extended declines in real terms of faculty compensation are not in the interest of the University or the Schools. The pool of funds available for faculty salary increases is awarded, however, according to merit, not as a cost of living adjustment. Since the financing of salary increases is a function of School budgets, there will continue to be variability in salary increases in different parts of the University.

b) As noted in the SCESF Report last year, Tables 2 and 3 (in the full report) give information about the percentage of faculty members receiving increases less than the rise in the cost of living, but they give data only for a single academic year. The real cost to the faculty member of a series of increases each of which is only slightly below the CPI growth percentages could be significant. In general, it would be useful to supplement Tables 2 and 3 with information cumulating increases and changes in the cost of living over a longer time interval. The Committee does not currently see such data and therefore cannot currently comment on whether or not this is a problem and, if it is, what the extent of the problem might be. The Committee asked last year to see such data in the future. The Committee wishes continued discussion with the Provost’s Office on appropriate frames for measurement.

The Provost’s Office agrees to explore this request with the Office of Institutional Research and Analysis.

c) In previous reports, the SCESF observed considerable variability in median faculty salaries across Penn’s 14 schools and areas. The Committee understands that both school and area finances and external conditions will inevitably influence such figures. Information about the extent of this variability and its course over time is nonetheless of ongoing interest. The Committee would like to continue to receive and analyze this data going forward.  

Variability in median-level faculty salaries across Penn’s Schools is, in part, a function of differences in salary levels among professional disciplines, as well as School budgets. The Provost’s Office remains committed to supplying school-level information to SCESF so that it can continue to monitor the issue of equity among Penn’s 12 schools.

3. Gender Equity
Data in Table 12 show that average salaries are lower for women than for men in all faculty ranks, even after weighting the data to reflect differences in the gender distribution of faculty by school and area. Persistent gender inequity in faculty salaries is troubling.  

SCESF Recommendation:
For two years, the SCESF has recommended that the Provost’s Office place priority on identifying the causes of observed gender differences in salaries and addressing any inequities that are not attributable to legitimate forces. In response, the Provost’s Office noted that the 2009 Gender Equity Report found relatively few significant differences by gender when years of experience, department, and school are considered. This analysis is incommensurate with the unweighted and weighted means and medians presented in Table 12. The Provost has commented that the issue would require further study. The SCESF requests a briefing on such further study as has taken place and concrete plans for the future.

The Provost’s Office is committed to the principle of gender equity in salaries. We are pleased to see that differences between men and women’s median base salaries are small and declining over time. Last year in response to the SCESF recommendation, we undertook a systematic investigation of gender differences in salaries in each of the Penn schools to identify the origins of individual differences. Differences in years in rank and other legitimate sources of salary variability accounted for the small, observed gender gaps in mean and median salaries. Going forward, we would be happy to discuss gender salary equity, the data in Table 12, and other information bearing in these issues with the Committee.

4. Faculty Benefits
As faculty benefits at Penn compared with peer institutions have not been thoroughly examined since the 1998-99 report, the SCESF appreciates the provision of data for Table 5 Detail 2and anticipates continued productive analysis of these data over time in complement to comparative mean salary data. Going forward, the SCESF continues to believe benefits should be comprehensively reviewed every five years to ensure competitiveness. 

SCESF Recommendation:
a) The SCESF appreciates the provision of data regarding mean total compensation this year and in the future. Analysis of these data offers a basis for some judgment of faculty benefits on an annual basis.

The Provost agrees that the provision of data on total compensation is a useful addition to the annual report on faculty salaries and will provide these data on a yearly basis.

b) The SCESF recognizes that examination of faculty total compensation incompletely analyzes faculty benefits within the university and across ranks. Sabbatical leave is among important benefits not addressed in current data. Consequently, we request that, in addition to data on mean total compensation, specific information regarding sabbatical leave within the university, across schools and compared to peers be provided where possible beginning next year.

The question of how the sabbatical leave benefit should be factored into the benefits package is complex. We agree to gather data on sabbatical leaves and to discuss how this data might be provided. 

Members of the 2011-12 Senate Committee on the Economic Status of the Faculty
William Dailey, SAS/Chemistry
Sarah Kagan, Nursing, Chair
Andrea Liu, SAS/Physics & Astronomy
Janice Madden SAS/Sociology
Tim Rebbeck, SOM/Biostatistics & Epidemiology
Petra Todd, SAS/Economics

Ex officio
Senate Past Chair, Robert Hornik, Annenberg
Senate Chair Elect, Susan Margulies, SEAS
Senate Chair, Camille Charles, SAS/Sociology

The Committee would like to explicitly acknowledge the essential and valuable assistance of Sue White of the Office of the Faculty Senate.