U’s response to national report on admin pay

We’re not unequal (emoeby via Flickr)

Having been named among the “Top 5 Most Unequal Public Universities” by the Institute for Policy Studies on Sunday, the University of Minnesota says the institute’s report — The One Percent at State U — is “outright wrong” and based on “inaccurate data” and a flawed methodology.

Here is the U’s updated response from this afternoon. The previous morning version comes afterward.

We consider this report extremely flawed because it uses inaccurate data and presents a misleading picture of the University. While the report addresses important issues in higher education – administrative costs, student debt and use of part-time teachers or adjuncts –President Eric Kaler and his administration have made significant progress on these issues. The report ignores this progress, and fails to address these issues in a responsible and truthful way for the years the report claims to represent. In some instances, the methodology and numbers that result are outright wrong.

The University’s analysis of the IPS report finds these significant errors:

1) All of the faculty and staff count data (“non-academic administration”, “permanent faculty” and “adjuncts”), are misrepresented because they are from different years than stated in the report, and the actual data for the years they claim to show tell a very different story.

The IPS report relies heavily on data presented by the American Federation of Teachers (AFT) on its website in a section called a Higher Education Data Center.  AFT indicates that all data “comes directly from the Department of Education, based on information reported annually by the institutions.”  This does not seem to be an accurate statement – AFT data often do not match Department of Education/IPEDS data at all, or matches data but incorrectly reports a different year.  These inaccuracies cloud the conclusions of the IPS study.

As examples, the report relies heavily on IPS’ definition of “Part-time/Adjunct Faculty.” We believe the IPEDS data (data the University reports to the federal government) that IPS attempted to capture are defined in IPEDS as “Part-Time, Instruction/Research/Public Service.” Below is a comparison of the two data sources as an example, showing how the same numbers appear in different years.  The accurate figures and years are those the University has provided to IPEDS. The University’s publicly available data, along with all institutions, is available at http://nces.ed.gov/ipeds/datacenter/. Note: We also believe AFT and the researchers freely confuse calendar year and fiscal year terms throughout their report.

AFT data (University of Minnesota)

Part-time/Adjunct Faculty Fall 2007 Fall 2009 Fall 2011 Fall 2012
    Men 335 289 712 744
    Women 137 118 574 574
TOTAL 472 407 1286 1318

IPEDS data – University of Minnesota Twin Cities

Change in IPEDS definition occurred between the 2006-07 and 2007-08 years.

Part-time/Instruction, Research & Public Service 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12
    Men 335 289 712 794 744 688 677
    Women 137 118 574 536 574 529 541
TOTAL 472 407 1286 1330 1318 1217 1218

For example, the AFT higher education data center shows a total of 1318 part-time/”adjunct” faculty in Fall 2012.  Those numbers are actually reported in IPEDS in the 2009-10 data submission, a full three years earlier.  Furthermore, the numbers cited are for any part-time employee in the instruction, research, or public service category, and so represent both faculty and non-faculty (e.g., lab scientist, extension educators, and so forth) in part-time roles.

In 2006-07, IPEDS changed the definition and captured more people in this category. Prior to that, the same people were present, but in a different category. Accordingly, it is misleading to compare data from 2006-07 to later years.  The reason is that between 2006-07 and 2007-08, IPEDS underwent a significant change in the definitions for reporting (i.e. more job codes were used to report the numbers, greatly increasing the number of these types of employees represented in the IPEDS data, even though the University saw almost zero employees added between those two fiscal years).

Setting the definition change aside, it would be a much better apples-to-apples comparison for one to consider the change in employees from 2007-08 to 2011-12. This shows that, in fact, the numbers of employees in this category have actually decreased.

The University’s own data on full-time vs. part-time faculty suggest a very different story than represented by IPS, showing a 3.7% increase in the number of full-time faculty, vs. a 3.6% decrease in the number of part-time faculty over the past six years:

Faculty Headcount – Twin Cities campus Fall 2008 Fall 2009 Fall 2010 Fall 2011 Fall 2012 Fall 2013
    Full-time 2,877 2,821 2,833 2,856 2,880 2,983
    Part-time 591 596 541 567 604 570
TOTAL 3,468 3,417 3,374 3,423 3,484 3,553

Source:  http://www.oir.umn.edu/static/hrdata/Employee_Head_Counts_2006_2013.pdf

2) The scholarship and fellowship expenditure per student grossly misrepresents the University’s student financial support. We believe IPS is using the wrong measure, but even so, we cannot accurately recreate the report’s claims based on the sources listed. The total amount of spending on scholarships has increased from 2005-06 to 2011-12 by 48.7%, and on a per student basis by 44.9%.

IPEDS data – University of Minnesota Twin Cities

Total gross scholarships and fellowships, & enrollment 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12
    Scholarships $151,061,675 $162,241,780 $179,319,015 $188,334,157 $218,229,110 $227,477,680 $224,848,435
    Fall Enrollment 51,175 50,402 50,883 51,140 51,659 51,721 52,557
Scholarship/enroll $2,952 $3,219 $3,524 $3,683 $4,224 $4,398 $4,278

The institution makes financial aid available to students in two ways.  One is providing a direct payment to the student that can be used for any purpose related to the expenses of attending the U, such as books, transportation, or other costs.  The second is to provide a discount or allowance applied to the tuition and fees billed to the student.  In 2005-06 the University provided $60.4 million in scholarship payments and $90.7 million in discounts or allowances, for a total financial aid budget of $151.1 million.  In 2009 we made an accounting change to more accurately capture and report financial aid that is provided directly to students as tuition discounts, and in 2011-12 we provided $186.5 million in discounts and allowances, and $38.3 million in scholarship payments, for a total financial aid budget of $224.8 million.  The FY06 to FY12 increase is $73.7 million, or 48.7%.  It appears the IPS lacks understanding of how financial aid is tracked in IPEDS, and neglected to note the strong increase in overall scholarship support.

3) There is no dramatic increase in non-academic staff. Setting aside the definition problems, if you simply look at the University’s headcount data from 2006 to 2013, the number of employees in the faculty category (both full- and part-time) rose by 10.1%. The number of all other employees excluding graduate assistants (both full- and part-time) rose by 0.4%.

Employee Headcount – Twin Cities campus Fall 2006 Fall 2007 Fall 2008 Fall 2009 Fall 2010 Fall 2011 Fall 2012 Fall 2013
    Faculty 3,226 3,390 3,468 3,417 3,374 3,423 3,484 3,553
    All other staff 13,477 13,637 13,938 13,479 13,399 13,267 13,370 13,529
TOTAL 16,703 17,027 17,406 16,896 16,773 16,690 16,854 17,082

Source:  http://www.oir.umn.edu/static/hrdata/Employee_Head_Counts_2006_2013.pdf

4) The expenditure data appears to be simply wrong. Lacking transparency in the report about how IPS derived its numbers, we were unable to recreate IPS expenditure data and determine why the numbers misrepresent reality. By the University’s calculations, the report’s “administrative” expenditures (institutional support) are more than $1,000 per student too high.  The University has undertaken a much more extensive annual study of its costs in all categories and presents those results to the Board of Regents and the public on an annual basis.

It is also important to note that while this report depicts President Kaler and his peers on the cover, the purported timeframe the report examines includes just one year under the leadership of President Kaler, and a budget he had nominal input in shaping. President Kaler began his U of M post in July 2011 (i.e., the start of FY2012). The report fails to acknowledge changes the University implemented in 2012 to construct a strict senior leader compensation and separation policy.

The report also fails to mention many other positive and very significant changes under President Kaler’s leadership, some of which include:

  • freezing tuition for all undergraduate resident students on all five campuses for two years;
  • a commitment to $90 million in administrative cost savings by the end of FY2019 (the FY2015 budget recommended to Regents last week represents nearly $36 million toward that commitment by the end of FY2015);
  • significant investments in student scholarships, including the Promise Scholarship (provides $30 million in aid to more than 13,000 students), the president’s Emerging Scholars Program and Retaining all Our Students;
  • nearly $100 million in investments between FY13-FY15 for key academic priorities, including recruiting and retaining field-shaping faculty members; and
  • a reduction in the size of President Kaler’s senior leadership team.

Overall, the report misrepresents the University of Minnesota, attempts to paint a very sensational picture to support IPS’ thesis, and ignores the reality at our institution. While there is still work to do, the University of Minnesota is making strong progress in providing access to students, holding down student debt, reducing administrative costs and investing in students’ future education and needs.

Here is the response from this morning:

We consider this report extremely flawed because it uses inaccurate data and presents a misleading picture of the University. We agree that the report addresses important issues in higher education – administrative costs, student debt and use of part-time teachers or adjuncts.

Since taking office in July 2011, President Eric Kaler and his administration have made significant progress on these issues.

However, the report ignores this progress, while also failing to address these issues in a responsible and truthful way for the years the report claims to represent. In some instances, the methodology and numbers that result are outright wrong.

The University’s analysis of the IPS report finds significant errors:

1) All of the HR data (“non-academic administration”, “permanent faculty” and “adjuncts”), are misrepresented because they are from different years than stated in the report, and the actual data for the years they claim to show tell a very different story.

As examples, the report relies heavily on IPS’ definition of “Part-time/Adjunct Faculty.” We believe the IPEDS data (data reported by the University to the federal government) that was attempted to be captured is defined in IPEDS as “Part-Time, Instruction/research/public service.” Below is a comparison of the two data sources as an example. The University of Minnesota’s publicly available data, along with all institutions, is available at http://nces.ed.gov/ipeds/datacenter/. Note: We also believe the American Federation of Teachers (AFT) data the researchers freely confuse calendar year and fiscal year terms throughout their report.

AFT data (University of Minnesota)

Part-time/Adjunct Faculty Fall 2007 Fall 2009 Fall 2011 Fall 2012
    Men 335 289 712 744
    Women 137 118 574 574

IPEDS data (reported by the University of Minnesota)

Part-time/Instruction, Research & Public Service 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12
    Men 335 289 712 794 744 688 677
    Women 137 118 574 536 574 529 541
TOTAL 1286 1318 1218

 

In 2006-07 IPEDS changed the definition and captured more people in this category. Prior to that, the same people were present but in a different category. Accordingly, it is unfair to compare data from 2006-07 because of the change in the definition of the category. When you look at the data after the definition change, it is apparent that the number bounces around a little bit, but an actual comparison of 2007-08 and 2011-12 shows a decrease of 68.

Though it does seem AFT and IPS attempt to use the IPEDS data, the misrepresentation of timeframe makes many of the report conclusions spurious. You will find similar data inaccuracies when examining any of the human resource data contained in the report. The reason is that – between 2006-07 and 2007-08 – IPEDS underwent a significant change in the definitions for reporting (i.e. more job codes were used to report the numbers, greatly increasing the number of these types of employees represented in the IPEDS data, even though the U saw almost zero employees added between those two fiscal years).

Setting the definition change aside, it would be a much better, apples to apples comparison for one to consider the change in employees from 2007-08 to 2011-12. This shows that in fact the numbers of employees in this category have actually decreased.

2) The scholarship and fellowship expenditure per student grossly misrepresents the University’s student financial support. We believe IPS is using the wrong measure, but even so, we cannot accurately recreate the report’s claims based on the sources listed. The fact with respect to scholarships is that the amount of spending has increased from 2005-06 to 2011-12 by 48.7%

3) There is no dramatic increase in non-academic staff. Setting aside the definition problems, if you simply look at the University’s headcount data from 2006 to 2013, the number of employees in the faculty category (both full- and part-time) rose by 10.1%. The number of all other employees – excluding graduate assistants (both full- and part-time) – rose by 0.6%.

4) The expenditure data appears to be simply wrong. Lacking transparency in the report about how IPS derived its numbers, we were unable to recreate IPS expenditure data and determine why the numbers misrepresent reality. By the University’s calculations, the report’s “administrative” expenditures (institutional support) are off by $1,000 per student.

In addition, much of the financial and human resource data used in the report is taken from data presented from the American Federation of Teachers website. In other words, the data is at least once removed from its origin (i.e., IPEDS data reported to the federal government directly by institutions), which opens it up to potential inaccuracies and leads readers to believe IPS’s report has political goals.

It is also important to note that while this report depicts President Kaler and his peers on the cover, the purported timeframe the report examines includes just one year under the leadership of President Kaler, and a budget he had nominal input in shaping. President Kaler began his U of M post in July 2011 (i.e., the start of FY2012). The report fails to acknowledge changes the University implemented in 2012 to construct a strict senior leader compensation and separation policy.

The report also fails to mention many other positive and very significant changes under President Kaler’s leadership, some of which include:

  • freezing tuition for all undergraduate resident students on all five campuses for two years;
  • a commitment to $90 million in administrative cost savings by the end of FY2019 (the FY2015 budget recommended to Regents last week represents nearly $36 million toward that commitment by the end of FY2015);
  • significant investments in student scholarships, including the Promise Scholarship (provides $30 million in aid to more than 13,000 students), the president’s Emerging Scholars Program and Retaining all Our Students;
  • nearly $100 million in investments between FY13-FY15 for key academic priorities, including recruiting and retaining field-shaping faculty members; and
  • a reduction in the size of President Kaler’s senior leadership team.

Overall, the report misrepresents the University of Minnesota, attempts to paint a very sensational picture to support IPS’ thesis and ignores the reality on our campus. While there is still work to do, the University of Minnesota is making strong progress in providing access to students, holding down student debt, reducing administrative costs and investing in students’ future education and needs.

The following provides greater explanation specific to the claims made in the IPS report.

Specific Claims

Claim #1:  Between FY2006 and FY2012, the University of Minnesota – Twin Cities increased spending on non-academic administration by 125% while decreasing spending on scholarships by 36%.  

We believe the researchers intended to use the IPEDS expenditure category of “institutional support” as a proxy for non-academic administration, though they used a derived variable of institutional support per full-time eployee. IPEDS expenditure categories vary for a number of reasons, and it is especially important to note that many of the system-wide administrative responsibilities are housed and reported with the Twin Cities campus, rather than a separate system office, as in the case of Wisconsin, Texas or California, for examples. The University has consistently argued that the use of the IPEDS institutional support expenditure category is an inappropriate proxy for administration, and now conducts annual system-wide studies and reports to the Board of Regents on our spending on mission, mission support and leadership, and oversight activities.

Correct data from U of M reports to IPEDS:

Institutional support per FTE, FY2006, $2,775 to FY2012, $4,702, increase of $1,928 or 69.5%.

Note: This is not controlled for inflation.  Using current dollars adjusted by CPI, this would be:
Institutional support per FTE, FY2006, $3,226 to FY2012, $4,779, increase of $1,553 or 48.1%.

The claim of decreasing of spending on scholarships by 36% is flatly false. The University has increased spending dramatically on scholarships and fellowships for both undergraduate and graduate students, both with significant fundraising campaigns and commitments to additional graduate fellowships, as well as our Promise Scholarship programs (provide $30 million per year in aid to low and middle income students).

The IPEDS reported total gross scholarships and fellowships for the Twin Cities campus was $151.1 million in 2005-06, and $224.8 million in 2011-12 (the most recent year available with final data), for a 48.7% increase.

Claim #2: From FY2010 to FY2012, the University of Minnesota increased non-academic administrative staff 200% from 762 to 2,384.

On its face, it is absurd to assume the University of Minnesota added 1,622 administrative staff in a two year period, when on average the University grows its workforce less than 0.3% in any given year. We believe the researchers attempted to use an IPEDS HR category labeled “executive/administrative/managerial” as a proxy for non-academic administrative staff, though again the timelines are not reconcilable. As noted above, in 2007, due to IPEDS data definition changes, many employees were reclassified in the IPEDS reporting. In particular, 3,501 employees were reclassified from a category labeled “other professionals” to different categories. While 1,633 of these employees were subsequently labeled as “executive/administrative/managerial”, an additional 1,797 employees were reclassified as instructional, research or public service employees.

The University has analyzed and reported these data reporting changes to both the Board of Regents and the State of Minnesota Legislature, and has also shown how similar institutions have had equal or greater changes in their reporting. The University annually reports to the Board of Regents a cost benchmarking study that more appropriately examines these issues. Note that IPEDS has moved away from these human resource definitions beginning in 2012 and has begun to use the Bureau of Labor Statistics categories for this reporting.

Using data from the correct years from IPEDS for this particular measure, the definitions are stable, and the change is from 2,396 in Fiscal Year 2010, to 2,481 in Fiscal Year 2012, an increase of 85 full-time headcount employees, or 4%.

We believe the report data is not accurate. Our annual headcount data shows virtually no increase.

Claim #3: While the president pulled in $2.1 million, permanent faculty decreased 9% and the ranks of adjuncts grew 223% to nearly half of all instructional staff.

For faculty numbers, we believe the researchers are relying on one of the several ways IPEDS classifies “full-time” faculty vs. “part-time” faculty as proxies for “permanent” and “adjunct” faculty. Though poorly worded, we believe the study attempts to use the FY2010 through FY2012 timeframe to make this assertion. In any case, the lack of clarity in both the data and timeframe makes the contention of changes in “permanent” and “adjunct” faculty meaningless (neither are defined as such in IPEDS). Additionally, several IPEDS data collection efforts around faculty exclude medical schools, which of course is a critical part of the University of Minnesota.

The University reports its summary HR data for all campuses publicly at http://www.oir.umn.edu/static/hrdata/Employee_Head_Counts_2006_2013.pdf. The data show a total faculty count for the Twin Cities campus in the Fall of 2006 at 3,226, and in the Fall of 2013 the same number is 3,553, for a 10.1% increase. This includes an increase of 179 full-time faculty and 148 part-time faculty.

The IPS report uses part-time staff who we have designated as being connected to the functions of instruction, research and public service as their proxy for “adjuncts.” Again, the report uses data linked to incorrect fiscal years provided on the AFT web site, which – in addition to being inaccurate – means the numbers are drawn from before and after a major change in the definitions, as described above.

Using the data for the correct fiscal years from IPEDS, the definitions are stable, and the numbers for this measure (part-time staff in the functions of instruction, research, and public service) are 1,318 for FY2010 and 1,218 for FY2012, a decrease of 100 employees, or a percentage decrease of 8%.

Claim #4: From FY2006 to FY2012, expenditures per student on non-academic administration more than doubled from $2,574 to $5,790.

Again, we believe the researchers are using “institutional support” as a proxy for non-academic administration on the Twin Cities campus, an assumption we believe is incorrect. We were again unable to reconcile this calculation with any institutional or IPEDS data, even trying to account for incorrect timelines. Many system-wide support expenses are recorded solely on the Twin Cities campus, as much of the administration and leadership serves both campus and system roles. For example, enterprise computing, technology infrastructure and system-wide offices for audit, legal and other services available to all of our campuses, extension offices and outreach centers are reported in numbers for the Twin Cities campus.

The increase in the “institutional support” category from 2008 to 2010 was due primarily to Governmental Accounting Standards Board (GASB) changes in the handling of post-retirement benefits, system-wide reconciliation of fringe costs and the accounting treatment of royalty streams, none of which were changes in “non-academic administration.”

The figures cited by IPS from the AFT website do not match those in IPEDS, and it is not clear what definition they might have used. Unlike with the employee headcount figures, it is not simply a matter of mislabeling the years.

For FY2012, the correct figure for institutional support per FTE would be $4,702. See additional explanation under Claim #1.

Claim #5: Meanwhile, expenditures per student on scholarships dwindled from $1,424 to $914. By 2012, average student debt reached $29,702.

As noted above, it is not true that the University has reduced scholarships, and the correct use of IPEDS to examine total scholarship expenditures would have given the researchers a more complete picture. On a per student basis, the figures listed in response to Claim #1 would translate to $3,643 in FY2006 and $5,130 in FY2012, an increase of $1,487, or 41%. Finally, while the average student debt figure cited by the researchers is at least close to our figure of $27,578, it is important to note that this is an average only for those students that have debt. Over one-third of University students leave with no debt at all, and the University of Minnesota continues to have the lowest net price of any 4-year school in Minnesota for low-income students.