A note on TPAC and tuition increases at UT Austin

In 2011 the Tuition Policy Advisory Committee of UT Austin recommended increasing tuition for in-state and out-of state students. The increases were shot down by the regents in a bid to make the republicans look more populist than Powers. A few weeks ago an email was sent out by our “student leaders” explaining that the TPAC recommendations would now be implemented. Tuition would now increase, except graduate students would be excluded. According to the “student leaders” the original TPAC recommendation was “supported by students”, but a 2011 referendum found that 64% of the student body was against the tuition increase [1]. Furthermore there is “no time” to have any formal process of feedback with the campus. Only a few student legislative body meetings will be used for town hall meetings before the recommendations are due.

There are a lot of issues to be discussed in relation to tuition and the budget, but in this article I comment on only one aspect of their argument which I viewed as extremely flawed. It is commonly stated that the amount of state appropriations that cover UT Austin’s total budget has shrunk dramatically over the years. A UT Austin website called “Budget 101” states that “In the 1984-85 fiscal year, the State of Texas provided 47 percent of the university’s total budget. For the 2010-11 fiscal year it has dropped to 14 percent.” [2] The TPAC committee is fond of presenting graphs such as the following which “demonstrate” the necessity to increase tuition.

tpac graph

The following graph was presented in 2011 with the caption “Tuition Increases have not fully offset declines in State General Revenue; efficiency improvements, administrative and programmatic reductions and other revenue enhancements have made up the difference.” This idea is intuitively convincing in that there were fairly large declines in the 2011-2012 biennial for state appropriations. However there is a severe problem with arguing that because state appropriations have declined as a percentage of overall revenue tuition needs to be increased to off-set that cost. First, the decline in percentage is partly (actually mostly) due to increases in overall revenue rather than a lack of state funding. In 1999 the overall revenue  for UT Austin was $1.36 billion dollars (adjusted for inflation to 2014), while in 2014 it is $2.35 billion dollars [3,4].

In fact if nothing else changed except tuition increased then we would expect the percentage coming from state appropriations in relation to overall revenue to decrease because increased tuition increases overall revenue. The TPAC argument is not only dishonest but viciously circular.

A more obvious way of comparing the relationship between state appropriations and tuition is by considering the real dollar numbers (adjusted for inflation) over time. The first thing one notices when one takes this perspective is that while state appropriations have decreased in recent years, they fluctuate over time in ways that suggest that they could increase again. According to Mr. Hegarty in an interview with the Daily Texan

“But they’re working on five-year forecasts under some very specific assumptions: One, that their revenues for the next five years that they have to spend are, at best case, flat. Why? You can’t deny 26 years of funding trends in funding from the state. Funding’s actually declining, not growing.” [5]

There has been a decrease in recent years, however, this statement is questionable at best. The following chart graphs inflation adjusted state appropriations over time (1986-2015). (Data are from the legislative appropriation bills available online [6]).

State appropriations

What these data demonstrate is that state appropriations were increasing overall from approximately 1992 until approximately 2005, but even then they only went down to 1996-1997 levels before the financial crisis. It is hard to detect strong trends because of the huge amount of fluctuations. Although there is a decline based a shortfall in appropriations in 2011 and 2012, the number has started to increase again. Furthermore, a decline is to be expected given the onset of the financial crisis in 2008-2009. According to a Grapevine survey of state appropriations in higher education 31 states increased appropriations a few years after the financial crisis. According to James Palmer, professor of higher education at Illinois state university; “Historically there have been ups and downs in state funding. The Great Recession created a huge trough, but barring further or unexpected declines, I think we’re beginning to climb out of the trough, at least at a national level”  [7, 8]. For example, in 1988 it seems there was a massive decline to $267 million (inflation adjusted to 2014). However in the coming years increases followed. The CFO is extrapolating based on too little data.

The argument of TPAC is that declines in state appropriations justify tuition increases. Tuition should compensate for a shortfall in state revenue. This argument is rendered problematic when one compares inflation adjusted tuition in relation to inflation adjusted state appropriations overtime for more than a few years [9]. The following graph charts growth in tuition revenue (red) and growth in state appropriations (blue).

appropriations and tuition

The CFO is not lying when he states that state revenues have decreased overall in the years, however the slope is extremely shallow and difficult to extrapolate from because of the high degree of fluctuation. A regression analysis shows that the decline in state revenue is not even statistically significant (p= 0.549).

Taking the CFO’s argument for granted, we find that state appropriations have decreased $564,000 on average each year. Here’s the problem with the tuition based argument. From the data, tuition and fee revenue has increased by an average of $15,435,000 a year. Tuition revenue has increased 27.4 times faster than state appropriations have decreased [10].

If one wanted to make an honest argument about the relation of state appropriations to tuition and fee revenue, one would also argue that when state appropriations increase, tuition should also decrease, but this has never happened. Historically, every fluctuation downward of state appropriations have been used to justify tuition increases, but the converse is never true [11].

Given the actual trends, the real question is not how do we recover state appropriations through tuition, but rather; Why does revenue need to constantly increase far above inflation? And why are these costs pushed onto students? If one is actually interested in lobbying for state funding, at the end of the day the party to convince is the tax payer. But an administrator that advocates tuition increases at every turn, thus making University less accessible to the public, is shooting himself in the foot if he is interested in acquiring more state funding, which will ultimately involve convincing the public.


[1] http://www.dailytexanonline.com/news/2012/03/02/university-sg-election-results-announced-tuition-referendum-also-attached

[2] http://www.utexas.edu/finances/money-from.html

[3] http://www.utsystem.edu/cont/reports_publications/RevExp/RevExpAnalysis.pdf

[4] http://www.utsystem.edu/cont/Reports_Publications/summaries/2014/FY14BudgetSummaries.pdf

[5] http://www.dailytexanonline.com/opinion/2013/10/21/qa-kevin-hegarty-on-shared-services-and-uts-declining-funding

[6] http://www.lrl.state.tx.us/legis/approBills.cfm

[7] http://www.insidehighered.com/news/2013/01/21/state-appropriations-higher-education-31-states-report-says

[8] http://www.sheeo.org/resources/publications/grapevine-compilation-state-fiscal-support-higher-education-results-fiscal

[9] Data for tuition revenue are from the following sources





[10] To a certain extent the increase in tuition revenue can be associated with an increase in enrollment. But not by very much. In 1988 enrollment was 50,107 students, while in 2003 it was 50,616. In fall 2013 the number was 52,059 (a decline from 2012). A more rigorous study would weight tuition increases in relation to enrollment, in addition to charting cost of living increases and other possible expenses.



[11] This phenomenon has been documented extensively see Ovetz (1996) and the sources cited therein.



Uninformed Student Senate President urges “informed discussion” on Shared Services at UT

Andrew Clark, the senate of college councils president, recently wrote an op-ed arguing that the debate on shared services has been “one-sided”. Clark makes the obvious point that a discussion of shared services should be “informed” [1]. Ironically, in spite of his argument, Clark’s article is riddled with factual inaccuracies and blatant mischaracterizations of the arguments from people who have raised concerns about the plan- a group of people Clark derides as “detractors”, without qualification.

Where is the evidence?

A case in point is the issue on the data for the Shared Services plan. Clark argues that the “detractors” are wrong to demand that the administration give data supporting the plan, because the data have not yet been produced, the whole point of the now in progress pilots. This mischaracterizes the position of the “detractors”. What has been demanded for the past year is the data referenced on page 19 of the Smarter Systems report from 2013 that were used in support of a Shared Services model by the Committee on Business Productivity [2]. This is completely separate from the data on the Shared Services that is supposed to be produced from the pilots. In the former case the data have been available, in contrast to what Clark states, they just had not been released to the public until a few weeks ago.

The data are now being analyzed by concerned students, faculty and staff. According to Mr. Hegarty, the actual data backing up the recommendation were of no concern to him. On October 30th he stated “I didn’t want the committee’s information, the committee didn’t offer it and we didn’t want it.” [3] He was asked the same question at the Graduate Student Assembly the same day and maintained that the administration had no interest in the data backing up the Shared Services recommendation [4]. In other words, they had no interest in assessing whether the CBP’s argument for Shared Services followed in any way from the data they gathered. The recommendation was accepted uncritically.

Clark’s narrative on Shared Services

Clark then constructs a paragraph long narrative on Shared Services in which nearly every sentence is false. He claims;

“Using professional expertise and information on UT, they made the determination that the University could save up to $30 million annually by centralizing operations like human resources, payroll, IT and procurement. Vice President and Chief Financial Officer Kevin Hegarty then assembled multiple committees made up of UT staff and faculty to examine the claim and verify its validity.”

This last sentence is completely false. The committees were not made to assess the claims of the Smarter Systems report regarding Shared Services. In fact the data from this report had not even been released. Even the cash-flow charts from the October Shared Services plan had not been produced when the aforementioned campus based committees were meeting. These were simply superficial information sessions, the primary purpose of which was to argue that Shared Services was not a form of “outsourcing”. Clark seems to have not even bothered interviewing a single staff, student or faculty member who went to the meetings he discusses (I was one). In fact the committee members were the same people who Clark derides for asking that data from the CBP on Shared Services be released. Clark then absurdly claims;

“It was determined that the figures were accurate, and so Hegarty spent the better part of the fall semester attending meetings and hosting town hall forums to engage as many people as possible on the project and solicit feedback from the campus.”

This statement is incorrect on multiple levels. Which “figures” were determined to be accurate? When these committees were meeting there were no “figures”, let alone the Accenture data to even back up the Shared Services plan. When the basic figures were released in the October 30th report it was found that the numbers from the cash-flow chart were off by $80 million and assumed that 423 staff members magically disappeared by December 2014. When the actual investment cost is put in the Shared Services plan does not apparently break even until year 10, if then. Graphs comparing the cash-flow chart from the October 30th plan and the adjusted cash-flow chart based on real investment are presented below [5, 6].


Clark then states;

“The data people seek is [sic] forthcoming, but it will not be here until units such as the College of Liberal Arts and the McCombs School of Business, both of which are part of a pilot Shared Services program designed to see how much the model could actually save, have completed their trial runs of the program.”

But Shared Services is not supposed to save anything until year 10, and that’s when its implemented at a campus-wide level. Clark’s statement is barely even coherent. The goal of Shared Services cannot just be to save money, it should be to save money in relation to the services provided, an issue that Clark does not even address (we could save a lot of money by eliminating the University of Texas). Furthermore, claiming that the sharing of employees between faculties in COLA and McCombs is anything like the massive over-haul that is recommended by the Committee on Business Productivity, with an off-campus call center, drastically reduced staff and the loss of department specific knowledge on a mass scale is a questionable position to say the least. That said, critical assessment of the Shared Services plan requires that the administration detail what evidence would actually cause them to not implement Shared Services after data have been gathered from the pilot. Until then the pilot is testing nothing.

Clark, who never attended a single town hall meeting or campus based committee on Shared Services, then concludes from his completely fabricated history that “The process has been reasoned, public and cautious.” I remind the reader at this point that the Committee on Business Productivity was composed of 13 business leaders, not a single one with teaching or research experience. Accenture was paid consulting fees up to $4.1 million to help produce the draft plan with obviously incorrect profit projections [7]. The Shared Services project team was composed mostly of Accenture consultants and not a single student or professor or non-managerial staff member. This committee decided to buy the ERP Workday during the summer of 2013 with no one around and no campus dialogue. At the second town hall meeting on November 13th Mr. Hegarty refused to allow follow up questions- all questions had to be submitted on a piece of paper beforehand and vetted. When an open question period was allowed there was not a single positive comment concerning Shared Services from the audience which included students, faculty and staff.

“Don’t throw the baby out with the bath-water” (Accenture=baby)

Clark argues that the critique of Accenture’s involvement is misguided. He argues that

“Just because some of Accenture’s employees may have made mistakes in the past does not mean the entire company is full of people who cannot be trusted to deliver on their contract.”

The issue is not that the company made some “mistakes”, but rather their poor performance in the public sector is almost systematic. When Accenture tanked food-stamp procurement in Texas, costing tax-payers $243 million, leaving 81,504 people (including children) uninsured and resulting in a five year backlog in the processing of food stamps [8] their profit margins did not suffer significantly. From the perspective of the company the botching of the food stamps was not a mistake, because there was no clear structure of accountability built into the plan. The actual worry is that UT’s intimate relationship with Accenture (head of IT Brad Englert, a former Accenture COO is on the project for one example), dramatically reduces the ability of UT administrators to act independently and in the best interest of the University. Systems of accountability both for Accenture and for UT central administration have not been spelt out, and the company was granted two contracts with the University without competitive bidding. The fact that Accenture helped draw up a plan that underrepresents costs in the order of $60-80 million should cause particular concern. Continued cost overrun in the public sector is endemic when public officials do not scrutinize claims made by consultants on the private sector [9].

Shared Services and the Cult of Efficiency

Clark then argues in favor of Shared Services by pointing out that it has been applied at other Universities. He argues that is has been implemented “[b]ecause it increases efficiency and thus lowers overall costs.” A noncontroversial definition of efficiency would be the quality of some service in relation to the cost. Efficiency is not necessarily increased through cost-cutting if the quality of service declines as well. Administrative staff at UT perform a number of non-identical and department specific functions. It is entirely possible that one of these services is adversely affected by Shared Services while another is not or one department is affected more than another. The Shared Services issue is not just a narrow question about saving money but touches on questions about the functions of a public institution of Higher Education. An assessment of University wide efficiency implicitly involves a weighting of the these different functions. Close attention needs to be paid to how efficiency is actually measured rather than narrowly focusing on how much money is saved. Questions such as “efficiency of what?, efficiency for whom?” need to be assessed [10].

To a certain extent we can assess the efficiency of Shared Services at the institutions that Clark mentions. While at both Berkeley and Yale there was a significant amount of disapproval of the plan, Michigan is the most important case since according to Hegarty “They’re probably the closest institution to us of any out there. They’re about the same size, the same scope, everything. And so we think: That’s really a good test case to look at, watch and monitor …. They look very, very much like we look when you look at how they’re organized” [11].

At Michigan faculty have complained that Shared Services dramatically decreased efficiency resulting in increased administrative work for faculty. The faculty council passed a resolution against the plan charging the following

“1. Reduction in faculty productivity by 10-20%.
2. Less faculty access to students and diminished quality of teaching for undergraduate students and supervision for grad students.
3.Loss of research funding to the tune of several tens of millions of dollars.
4. Increased frustration and consternation by the faculty because a significant fraction of their effort is diverted into secretarial-like tasks.
5. No cost savings; on the contrary, a great deal of loss in revenue.
6. Dehumanization of some 300 staff members.” [12]

The implementation of Shared Services at the University of Michigan was so controversial that the leader of the project and CFO, Rowan Miranda, had to step down as the leader of the Shared Services initiative [13]. A month later, the Michigan Daily reports that Miranda left his job at the University of Michigan [14].

At Michigan, Shared Services was predicated on cutting 50 jobs. The plan at UT is based on downsizing to an unprecedented degree, ten times as much as Michigan (with the elimination of 500 jobs) with a huge investment cost of at least $160 million. Any argument about “efficiency” needs to take into account the global consequences of such a plan, rather than hand waving about the amount of money saved through cuts.

“The University’s hard-working staff”

The most disingenuous argument that Clark makes is about the need for the implementation of Shared Services to be run by “the University’s hard-working staff members”. He derides the Texas State Employees Union “orchestrated” protest against Accenture’s involvement as an example of the “one-sided dialogue”. TSEU is composed partly of staff workers who have bravely been speaking out against the plan. Clark’s entire article is dismissive of these people. At a minimum he should have pointed out that there are disagreements between staff.

The comment is not just disingenuous because of the fact that it brushes over real conflict, it is also completely at odds with what is required of a Shared Services plan. In a section titled “Don’t Forget the Lawyers”, Bergeron in his book, Essentials of Shared Services explains

“Given the likely resistance to the shared services model by some employees, the Human Resources and Legal Services divisions of the corporation are likely to be hard hit with new work especially initially” [15]

Bergeron goes on to clearly articulate the attitude that management must have towards its employees in the implementation of Shared Services;

“As virtual pawns in the shared services implementation game, employees of both the parent corporation and those on the payroll of the shared business unit typically experience a high degree of volatility. This upheaval in normal operations is greatest early on, when employees associated with the back-end services are destined to be moved to a shared business unit and either downsized or retained in a new highly competitive environment.” [16].

Clark points out that “most” of the downsizing will be due to “natural attrition”, which implies that some percentage will be laid off (he has no idea how much). He believes that Shared Services will save jobs, but is unable to articulate why. Staff who raise concerns about the plan risk being terminated and Clark’s patronizing attitude does nothing to address the concerns of those staff he derides as “detractors”.


In addition to writing a breathtaking amount of falsehoods in his article Clark argues that the dialogue on Shared Services has been “one-sided”. Why? Because some people disagree with the administration. We should stop using “charged rhetoric” about “data and Accenture”, and engage in “informed discussion”. One wonders how informed discussion is supposed to take place without data, and when “student leaders” such as Clark refuse to do basic homework on the issue. Clark mysteriously does not cite any of his sources so it is unclear where he got his story on Shared Services from …. but I have a pretty good guess.

'We pride ourselves on being open to any new ideas the boss has.'

[1] Andrew Clark “Shared Services dialogue has been one-sided”, The Daily Texan February

[2] Committee on Business Productivity, “Smarter Systems for a Greater UT,” Jan. 2013.

For more detail see Villarreal and Tallman SS Draft 11

[3] Podcast of the Townhall meeting; Hegarty makes that exact statement at approximately 39:00
[4] For the GSA podcast please the following link. Look for the meeting on October 30th
[5] Alberto Martinez, “Should we centralize staff at UT?” Presentation before the UT Graduate Student Assembly, December 13, 2013.

[6] Villarreal and Tallman SS Draft 11

[7] http://www.utexas.edu/transforming-ut/committees/administrative-services/shared-services-committee

[8] Martinez “UT’s relationship with Accenture should raise questions” http://www.dailytexanonline.com/opinion/2014/01/20/uts-relationship-with-accenture-should-raise-questions

Rahamatulla, Altaf. Texas Again Demonstrates the Pitfalls of Privatization.Progressive States Network, March 18, 2010,

[9] Cited in Alberto Martinez’s “UT’s relationship with Accenture should raise questions” The Daily Texan
http://www.dailytexanonline.com/opinion/2014/01/20/uts-relationship-with-accenture-should-raise-question. These statements were made at the townhall meetings available here; http://www.utexas.edu/transforming-ut/news-events

[10] A detailed critique of notions of efficiency used in higher education can be found in Welch, A. R. 1999. “The Cult of Efficiency in Education: Comparative reflections on the reality and rhetoric.” Comparative Education 11:47.

[11] http://um-openletter.eecs.umich.edu/

[12] Villarreal and Tallman SS Draft 11

[13] “University chooses new leader for shared services initiative”, The Michigan Daily, December 11, 2013.

[14] “Finance VP and former shared services leader will return to Chicago”, The Michigan Daily, Jan. 21 2014.

[15] Bryan Bergeron, Essentials of Shared Services (Hoboken, NJ: John Wiley & Sons, Inc., 2003), 23.

[16] ibid: 190-191.