Financing the University – Part 23


Charles Schwartz, Professor Emeritus, UC Berkeley, March 20, 2013




Part 22 of this series presented my latest calculation of the actual cost for the University of California to provide undergraduate education: it came out to an average cost of $6,910 per student for 2011-12.  The current level of mandatory fees for resident undergraduate students is $13,181, which is almost twice that amount. I invited readers to respond with comments and criticisms on this controversial topic.



Email to     March 5, 2013

Subject: True Cost of Undergraduate Education

Dear President Yudof;

Last week I issued a new paper, Financing the University - Part 22, with the subtitle, "Cost Accounting at a Research University." This is posted at  and I also mailed printed copies of the paper to every member of the Board of Regents, UC Vice Presidents and Chancellors, and leaders of the Academic Senate.

At the end of that paper I invite all readers to respond with their own comments and criticisms; and the purpose of this note is to ask you, as President of this great public university, to encourage your colleagues in the upper realms of the Administration to engage in that response.

This paper contains my latest calculation of what it actually costs this University to provide undergraduate education; and, when that number is compared with the current level of student fees, one sees that our undergraduate students are, on average, being substantially overcharged. While it used to be said that state funds subsidized undergraduate education at UC, now it is true that undergraduate tuition is being used to subsidize other missions (like faculty research and graduate programs) beyond the full cost of undergraduate education.

You are familiar with this analysis, since I have tried on a number of earlier occasions to engage you and your office in serious discussion of this problem.  The faulty accounting practice of hiding the cost of faculty research under the heading of Instructional Expenditures is not unique to UC; but the fact that all research universities engage in this financial subterfuge does not make it less dishonest.

I hope that the dedication to open and rational debate, which we so often cite as the cornerstone of university purpose, will lead to a rich discourse on this issue, with contributions from all over. Please accept my invitation to participate.

Sincerely yours, Charles Schwartz

I. No Responses from on High


     I have received no responses from any administrative officials of this University.  Any one of them might have responded to my accusations of their incompetence (regents) and dishonesty (executives).  Any one of them might have responded to my challenge: justify the NACUBO accounting practice, which they rely on. Any one of them might have offered some criticism of my methodology for calculating the cost of undergraduate education. Does this silence mean that they find their own behavior indefensible and accept my calculations of the true cost of undergraduate education?


II. Conceptual Responses from Faculty Colleagues


IIa. I think writing down costs without assessment of quality of the product they pay for are not useful ... I think you have to ask the question, suppose you completely removed research from UC, would the quality of the education we offer be the same?


     This is a common case of confusing two distinct concepts: What does it cost to produce a product or provide a service? What is the value of that product or service to someone who might want to buy it? The first is a straightforward problem in accounting (called “cost analysis” when the producer has several products or services); the second is a highly subjective judgment, which different people may see in quite different lights.


     My study is only about the cost analysis. I think this is a necessary first task, without which one cannot sensibly proceed to consider what the price should be.


IIb. I for one wouldn't teach at this University if I couldn't pursue my research here. And I believe my teaching is better because I use examples from my research in the classroom. ... I don't view these aspects of my work as separable and I don't want to work somewhere that does view them as independent.  


     Again, misconceptions. I have no intent or desire to separate faculty teaching from research. Where did you get that notion?  Of course, there is some enrichment from one to the other (just how much is open to debate); but does that mean one cannot ask the question about how much the University actually spends on each of its multiple missions?  The University does offer its own answer with its published calculation of “The Average Per-Student Cost of Education”. And what they say is that, in effect, the only purpose in hiring faculty is to provide superior education to its students (implying  undergraduate students). That means that faculty research – what we professors do throughout the academic year while paid our academic salary – has no other purpose!


     An example of this confusion between the concepts of “cost” and “value” may be found in the 2002 report from NACUBO (the National Association of College and University Business Officers), which set the accounting standards at issue here.  Their document is titled, “Explaining College Costs: NACUBO’s Methodology for Identifying the Costs of Delivering Undergraduate Education.” We find, on page 9:

 This methodology was created to help individual institutions calculate the annual cost of providing an undergraduate education. .... Nor should the cost information derived from its use be construed as a measure of the value or quality of the education provided by the institution. NACUBO never intended its Cost of College Project to address issues of value or quality, “


     Then, when discussing Departmental Research (page 27), they turn around and say,

“Several alternative proposals were considered, but NACUBO concluded that all departmental research costs should remain within instruction and student services. Departmental research is vital and has a direct impact on the value and quality of instruction provided to students.”


IIc.What happens when you apply your methodology to institutions with a different mission than UC? What is the cost of undergraduate education at CSU [California State University System]? What happens when you apply your methodology to a 4 year liberal arts university whose sole mission (it could be argued) is undergraduate education? The magnitude of tuition [there] is typically $40,000.


     Here is some data from CPEC for 2005-06.  Per-Student Cost of Instruction at UC is $18,203;  at CSU it is $11,624; at CCC it is $5,461.  My number for UC is lower than that for CSU. How is that possible? Faculty at CSU carry about twice the course load of the average professor at UC; they do not have graduate assistants as we do, to save them time from auxiliary classes and office hours and grading papers, etc.


     We, at UC, provide a very cost-effective and high quality undergraduate program. We also provide world class graduate programs and faculty-led research. The total cost for all of that is rather high; but the portion of that cost allocable to the undergraduate education mission is rather small. Hooray for us!


     If you compare UC, or any public research university, to a quality liberal arts college, then there is a huge cost difference. There is also a huge difference in what the students get in terms of personal attention from faculty who are devoted to undergraduate teaching as their first priority. Different students need or want different types of educational environment; and that can imply very different per-student costs.


III. Technical Responses from Faculty Colleagues


IIIa. [Your] estimate of instructional cost depends in part on a time-use survey that suggests that UC faculty devote 21% of their time to teaching.    I think this estimate is low as a UC average: it may be two or even three times higher for ladder faculty outside of the natural sciences and engineering fields that have lower teaching loads to reflect their obligations to their sponsored research.


     The survey covered all full-time I&R faculty throughout UC (excluding Health Science and Law) and all I needed was the average time spent on the various job-related activities reported.  Indeed, there are large differences in “teaching loads” among the various academic disciplines: it may be as high as 5 or 6 quarter courses per year in the Humanities and as low as 2 quarter courses per year in the Biological Sciences.  I encourage people to look at the actual report from the time-use study, which I have posted at  On pdf page 84 is a graph showing the distribution of reported total time spent on all UC-related activities over a 2-day period. This is a fairly compact distribution; and the average gives their result of 61.3 hours-per-week. On pdf page 83 is a graph showing the distribution of time spent on all Instructional Activities over a 2-day period. The average works out to be 26.0 hours-per-week; but this is a very broad distribution. Different individuals carry very different teaching schedules, even varying in time as well as by discipline; but all we need is the average.  There is further data, showing a 50-50 split on average between undergraduate and graduate courses. This gives a primitive answer for my costing analysis: (1/2) x (26.0)/(61.3) =  0.21; that is 21% of faculty work time, on average,  spent on undergraduate teaching (and that includes many aspects of teaching: course preparation and student advising and letter writing and committee meetings, etc., in addition to time in the classroom.) Then, there is further data where faculty specify that some time intervals reported as primarily devoted to research or service also contributed to instruction; and then I also make an allowance for sabbatical leaves. See my 2007 paper, posted at  for details of these fine points.


     The final 21% factor is then applied to get the Professors’ component of the direct cost of undergraduate instruction. But first I must take out the cost for Lecturers and Graduate Student Instructors, which is almost entirely devoted to undergraduate instruction.  These two components – professors and other teaching staff – turn out to contribute almost equal amounts to the total direct cost of undergraduate instruction.


IIIb. What about the cost for departmental support staff?


     In my 2007 paper I wrote that this is perhaps the greatest point of uncertainty in my calculation. There is no data available telling how their work time is distributed between the different missions. According to the official UC Budget, General Campus Instruction funds go 60% for teaching staff and 38% for supporting staff. My simplest guess was that this non-academic staff is deployed in the departments to meet the needs of the professors; and so I used the same 21% figure here.  But some colleagues have questioned this as being seriously inaccurate.


     I have now found another source of official data (“Departmental Allocations”) that shows a much smaller proportion of staff salaries, compared to academic salaries, in the budget for Instruction.  So, this is a modest question that needs to be resolved – if only the knowledgeable UC officials would deign to engage in discussions.


IIIc. What about Academic Support; isn’t this mostly for undergraduates?


     Aside from Libraries, the cost of which I allocate 1/2 to undergraduates, the main components here are computer support, Deans’ offices, and a variety of research and medical facilities and services (“Ancillary Support”), and I allocate ¼ of that cost to undergraduates. Again, some more specific data would be helpful here.


IIId. What about overhead costs? Federal grants allow about 50% as an overhead charge.


     I am not clear on how those external overhead rates are calculated; they undoubtedly involve a lot of cost at the level of departments and research units. My focus here is on the campus-wide or system-wide level of administration and facilities costs, which the University accounting system calls Institutional Support and Operation and Maintenance of Plant. My estimate of 11% overhead comes from spreading those (IS+OMP) expenditures over all (or not quite all) operational expenditures. I notice that UCOP (University of California Office of the President) does a similar thing – levying a uniform tax on each campus according to its overall expenditures – to fund its own administrative functions.


IV. Outside-the-Box Responses from Faculty Colleagues


IVa. I think the figure on undergrad teaching fraction of time is probably lowish … it is only fair to compare against a 40-hour workweek when accounting for the fraction of our “paid” time spent on teaching and teaching-related stuff.


     This is a cute argument but it is false. Professors at UC (and probably elsewhere) are not hired for a 40-hour workweek. Quoting from University of California Regulation No. 3, dating from 1935 [Academic Personnel Manual, Policy 005]:

“The Senate assumes that each of its members is devoting all his time and energies (his full ‘working’ time) to the University. Such service to the University includes varied types of activities, such as classroom teaching, conference with students, studying and writing, research, committee work, administration, and public service. Members of the Senate who are not engaged in certain of these activities will naturally have more time for others.”


IVb. Should the university engage in price discrimination based on "ability to pay" wherein some undergrads are asked to pay far more than others, on the basis of their family's wealth+income?


     This is about the UC policy and practice of taking 1/3 of student fees/tuition and dedicating it to need-based financial aid. As stated before (in Part 22), neither UCOP nor I consider any part of student financial aid as a part of the cost of education. Nevertheless, this is a difficult and important question.  Ideally this should be an obligation of the general society, with tax revenues on the federal and/or state level used to provide access to quality public education (at all levels) to all qualified students, regardless of their financial ability to pay. UC now supplements Pell Grants and Cal Grants with this 1/3 return-to-aid policy and most think this is a good thing. Yet, one can raise the question of whether The Board of Regents has the authority to levy what is, in effect, a selective tax; and beyond legalities, Is it right to do so?


     Personally speaking, I say, Yes. But I can understand that other citizens may see it differently. I also remember the story of Proposition 209, which ended UC’s affirmative action programs, which were well intended to achieve the social goal of equal opportunity. Therefore, as a matter of principle, I would encourage efforts to return the obligations for student financial aid entirely to the public/political realm.


     The calculations being discussed in these papers are relevant to this issue. Earlier, one could say that no students paid the full cost of their education at UC because the State provided a significant subsidy; thus, an additional “tax” might be justified. But that is no longer a true assessment of the financial flows.


V. Noises from Sacramento


     My thanks to Bob Samuels for bringing the following two items to my attention. 


From the 2013-14 Governor’s Budget – Omnibus Education Trailer Bill, Sec. 59:

“Require UC to Report on Cost of Education, Broken out by Undergraduate and Graduate Instructional Costs and Research.”


There was a response from UCOP to a recent similar move in the State Legislature [Letter from UC President Yudof to Senator Mark Leno, October 3, 2012]:

“ … The University has consistently stated that information on cost of education by level of student – or expenditures by level of student – are impossible to determine, given the myriad way in which funds provided from the State and other core funds are used. For example, the largest cost driver in a cost of education calculation is faculty salaries. When faculty are paid for their work at UC, they are being compensated for teaching, research, and public service. They are not paid for these activities separately, and the degree to which each faculty member participates in each activity varies by term and by year. Moreover, it is virtually impossible to separate out research costs from teaching costs at the graduate level – so much of teaching in graduate programs occurs through the research program.“


     I agree with what UCOP says here about research and teaching in the graduate PhD programs. That is why, in my own analysis, I do not attempt to make that separation but simply note the package of costs labeled “Faculty Research (unsponsored) plus related Graduate Programs.” Moreover, I see no need for making such an inquiry since there really is no issue about the fees being charged those students: graduate students in the PhD program at any high level research university invariably have all their fees covered by fellowships or by fee waivers given as a matter of policy.


     My focus is on Undergraduate Education. And I notice that the UCOP statement above does not give any reason why the costs of undergraduate education can not be separated from other costs within the Core Funds.  Indeed, the method of Activity Based Costing is the standard way of making such a separation; and the Faculty Time Use Survey (conducted by the University) provides the key data for that calculation. So, it will be interesting to see how this new initiative from Sacramento plays out.