March-April 2011

by Elizabeth D. Capaldi and Craig W. Abbey

Although virtually all universities have accounting systems that track every dollar in accordance with the accounting rules required by auditors, these systems are not adequate to inform their leaders or the public about the profit and loss, productivity, or efficiency of their activities. Instead, they obscure the different revenue sources on which universities rely, the true costs of core functions such as teaching and research, the cross-subsidies of some functions by others, administrative growth or shrinkage, and differences among disciplines. Without this information, institutional managers simply cannot make wise decisions.

Current financial systems are designed to ensure that funds provided to a university for a specific purpose are used for that purpose, not to provide accountability for performance. As a result, there are few useful ways to acquire information about the cost effectiveness of various activities. This separation of performance-related information from financial data prevents a clear understanding of the relationship between the income, expenses, and results associated with the production of teaching, research, or other university undertakings.

Given the current financial crisis facing states and individuals (who between them are the major funders of public institutions), colleges and universities need to focus on improving performance within their available resources while maintaining quality. The lack of good data hampers their ability to achieve these ends and interferes with the ability of states, boards, and other constituencies to understand their performance.

Moreover, to the extent there are measures of the productivity of teaching, research, and other functions of universities, they are not reported in a standard way. Without good and comparable data, there cannot be meaningful benchmarking of performance.

In this article, we describe some of the important information that gets buried in higher education budgets and suggest ways to present data that would be both useful for managing to maximize efficiency—and thereby generating revenue to maintain and increase quality—and comprehensible to external constituencies that are trying to understand the performance and effectiveness of universities.

The Operating Budgets of Universities

The operating budget of a university is normally not meaningfully articulated, even within the university. One reason is that the business officers of universities prepare financial reports for auditing purposes, not for use in managing universities.

Worse, the reports give a distorted view of the institution's revenues and costs to boards, legislatures, and the public and provide no information on what might be appropriate measures of quality and productivity. Without such information, those with little or no higher education experience have difficulty understanding the basic financial structure of the academic enterprise.

Fund Accounting

For colleges and universities, fund accounting is the norm. While for-profit enterprises use a single ledger, higher education institutions use multiple funds to meet various financial reporting requirements, and their financial statements follow the fund-accounting model. IPEDS (the Integrated Postsecondary Data System), tied to the audited financial statements, is based on the same system.

Funds are assigned to some pre-existing institutional entity; on the academic side, it is usually the academic department. The entity's expenditures are treated as a block, and whatever its major function is defines the fund. Academic departmental expenditures are thus typically counted as instructional expenditures, since that is the dominant function of the academic department.

This muddles the data in a number of ways. First, in a research university faculty teach, conduct research, and do public service. The time spent on research and the costs of that faculty time should be attributed to research, not instruction.

Imagine a faculty member paid $100,000 with a 60 percent assignment in teaching, 30 percent in research, and 10 percent in public service. The entire $100,000 will be counted as having been spent on instruction, when in reality $60,000 was spent on teaching, $30,000 on research and $10,000 on public service.

Even worse, the costs of academic administration (department chairs and staff) are included in the departmental budget. This means that in financial reports, the costs of academic administration in the departments are included in instruction, not in academic administration where they belong. These costs are real and can be substantial (Capaldi, 2009).

Further, some institutions pay faculty who serve as deans or associate deans out of their home departments' budgets, while others pay them out of a central decanal budget. If they are paid from the department budget, they show up in instruction; if they are paid out of the decanal budget, they show up in administration. In some decentralized universities, these costs may show up in different buckets for different schools or colleges. A simple change in the fund these costs are attributed to can lead to what looks like growth in academic administration even when nothing has really changed.

Revenues and Costs: Instruction

Revenue Sources

As well as having many functions, universities have many sources of revenue. But their annual financial reports do not differentiate revenues by source, and the diversity of revenue streams makes understanding an institution's budget difficult, even for those, such as knowledgeable board members and legislators, who are familiar with higher education's varied purposes and unique rules and regulations.

For example at public research universities, state appropriations constitute anything from virtually 0 percent to 79 percent of total revenue, with a typical university at about 25 percent. The other 75 percent is made up of tuition and fees, grants and contracts, auxiliary services such as residence halls and food services, and hospital operations (if the institution has a hospital). Boards and others can misinterpret these data because in the businesses they know there are not multiple sources of income that must be accounted for separately.

Boards are most interested in undergraduate education. But in the financial reports of colleges and universities, auxiliary services (e.g., residence halls, food services, student health services, and intercollegiate athletics) are included, making universities look much richer than they are. While these business-like operations are self-supporting (except for athletics, which in only rare instances pays for itself), they generally do not generate substantial extra money that can be spent on instruction or other academic functions. Including auxiliary income in the total amount of revenues distorts the amount of money available for education.

Revenue and Costs by Function: Students

Students are a source of revenue; at the same time, each additional student also generates expenses. But annual financial statements do not include enrollment. When student numbers go up, revenue goes up, but revenue per student FTE does not necessarily also rise. When looking at absolute revenue numbers, a board member cannot tell how much is available per student FTE, which is the number that informs them about the quality of instruction possible and the efficiency of the academic operation.

For instance, in FY 2008, Arizona State University had a total of $918.7 million in total funding from the state, plus tuition, for $15,175 per FTE. In FY 2010, ASU had a total of $977.9 million in tuition and state allocation, but the student FTE had grown from 60,543 to 66,988, so the per-FTE value had declined to $14,598. This per-FTE value decrease is reflected nowhere in the financial statements.

Costs by Discipline

Costs vary by discipline, another fact that is not revealed in normal financial reporting. While larger classes reduce instructional expenses, not all information can be learned well in a large class. For example, learning a musical instrument requires one-on-one instruction, clinical instruction in medicine or nursing permits only a small number of students to accompany an instructor on rounds, and scientific laboratories are best taught with smaller numbers of students than lectures. Assuming uniform tuition for each program, as is the case at most universities, some disciplines make and some lose money. Net revenue generators are typically the humanities, education at the undergraduate level, business, and pharmacy. Money losers are science and engineering, nursing, medicine, veterinary medicine, and the fine arts.

In one form of university budgeting, responsibility-centered management, revenues are theoretically returned to the unit that generates them. In this system, money must be taken off the top to subsidize the money losers, which cannot bring in enough revenue to cover their costs. When undergraduates pay the same tuition for all programs, money must be taken from programs that are inexpensive to produce (e.g., the humanities) and given to those that are more expensive to produce (e.g., science and engineering).

This basic financial structure of cross-subsidization is hidden in financial reports, invisible to those who call for supporting science and engineering—to generate institutional prestige and for purposes of economic development—and minimal investment in the humanities. But the STEM disciplines are subsidized by students taking English and history courses.

Figure 1 shows the tuition generated (revenue) minus budgeted expenditures at ASU for 2010, by college.

The basic relationships shown in Figure 1 are not unique to ASU. The data for the University of Florida and SUNY Buffalo are essentially the same, with the humanities, teacher preparation, business, and the social sciences the main profit makers and the sciences and engineering the big money losers.

Caption: Figure 1. Tuition Less Budgeted Expenditures At ASU, 2010

As mentioned above, in responsibility-centered budgeting each college is able to keep the money earned. So in theory, responsibility-centered budgeting encourages units to generate more of their own money, since they can keep those resources.

However, a system in which every college keeps every dollar earned by teaching creates a tremendous incentive for colleges to duplicate courses taught in other colleges in hopes of gaining a marginal benefit. Colleges of business will teach English and math, while colleges of engineering will offer physics courses. This practice duplicates expensive facilities and faculty. Most universities that use some form of responsibility-centered budgeting create countervailing incentives or other mechanisms to avoid this undesirable internal competition.

Differential Tuition to Deal with Differential Costs

Another way to deal with the discrepancies in costs between disciplines is to charge a differential tuition rate by discipline. In most universities, the rate provided per undergraduate student credit hour (whether from tuition or state subsidies) is uniform across academic programs and all levels and types of undergraduate instruction. This strategy may provide for ease of administration, but as mentioned before, it also results in very substantial internal subsidies from students in lower-cost programs to students in higher-cost ones.

Some undergraduate degree programs (for example, engineering) virtually guarantee a high income immediately after graduation, and there is less concern about saddling these students with debt they will not be able to repay than there is for those students headed for less lucrative careers. Graduate and professional tuition already reflects this judgment. Medical school tuition is appreciably higher than that for other graduate/professional degrees for two reasons—medical education is very expensive to provide, and recipients of medical degrees are guaranteed a substantial income after graduation.

It is perhaps not surprising then, that charging higher tuition for undergraduate engineering programs than for other programs is increasingly common in higher education; it is a program that is expensive to produce, and engineering graduates are in demand for high-paying jobs. Table 1 shows a sample of the differential tuition charged for engineering in selected institutions.

Table: Table 1. Differential Engineering Tuition and Required Fees for Selected Schools, 2010–11

There are two primary objections to charging differential tuition by program. First, many believe that students majoring in different programs but attending the same classes should be paying the same tuition, regardless of their major. To deal with this objection, many institutions charge differential tuition only for upper-division work (courses in the major) and a common tuition for lower-division enrollment, when general education courses are taken.

Secondly, there is a fear that some lower-income students might select a major because of lower tuition and not because it is the appropriate major for them. In response, universities that have instituted differential tuition generally return some of the increased revenue to financial aid.

Costs by Level of Instruction

Costs also vary by level of instruction. Lower-division education is less expensive than upper-division, which is less expensive than non-thesis master's education, which is less expensive than thesis master's, which is less expensive than doctoral education. The profit generated by the humanities and social sciences is due in part to lower faculty salaries (including a greater use of adjuncts), but another important driver is the general education curriculum—which requires all students to take arts, humanities, and social-science courses—and the fact that it can be taught relatively cheaply and often in large classes. Because financial reports do not show lower and upper division separately, confusion and poor public policy can result.

Lower division is particularly low cost to teach at big public research institutions, where teaching assistants help with the labs and recitations and lecture-class sizes can be very large. Upper division, in contrast, involves smaller classes with more specialized equipment and faculty and thus greater expense. Not understanding this, many legislators think community colleges are a less expensive way to educate students than universities and suggest that community colleges should teach courses at the upper division.

But community colleges teach only lower-division courses, so the appropriate cost comparison is between community colleges and the lower-division cost at universities. Most financial reports do not show the cost of lower division separately from upper division, so it is not surprising that there is confusion. But Florida does show these costs separately in its annual expenditure analysis by level. Figure 2 shows the average across all the universities in for expenditures per FTE for lower- and upper-division enrollments (see http://www.flbog.org/resources/iud/expenditure_search.php).

Caption: Figure 2. Lower-vs. Upper-Division Costs at Florida Public Institutions, 2009

Generally speaking, the larger the university in Florida, the lower the cost of teaching at the lower division, as shown in Table 2.

Table: Table 2. Expenditures per Lower-Division Credit Hour, 2009

A recent paper (Romano & Djajalaksana, 2010) from the Cornell Higher Education Research Institute compares the costs of lower-division study at community colleges and universities and concludes, with some caveats, that it is generally less expensive to educate lower-division students at universities. [Editor's note: see the article by Peter McPherson and David Shulenberger in the January/February 2010 issue of Change, which argues the same point.]

Because of the cost differential, some universities charge more for upper-division than lower-division enrollment. The University of Michigan charges differentially both by program and level as shown in Table 3, in accordance with the cost differences of the programs and the likelihood of highly paid employment immediately following graduation. The university also monitors access to programs and adjusts financial aid to ensure that access is not affected by the pricing plan.

Table: Table 3. University of Michigan Tuition per Semester by Division and Major, 2010

Revenues and Costs: Sponsored Research

Sponsored research is both a source of revenue and a cost. Each university reports the annual awards and the yearly expenditures of these funds separately, and most publicity highlights the revenue. But research is very expensive, requiring facilities, highly paid faculty, computing resources, and libraries way beyond what are required for the teaching mission—not to mention effort reporting, IRBs, IACUCS, hazardous waste disposal, etc. Animal labs require 15 air changes per hour and backup electricity sources; sophisticated equipment in research facilities requires stabilizers, sound proofing, and other hugely expensive features.

While each university negotiates with the federal government for the indirect costs it can bill on grants, the government does not allow billing for the total costs, and the university is unlikely to see even the amount that has been negotiated. So at ASU for example, the documented indirect cost rate is about 63 percent, the indirect cost ASU can bill on grants as negotiated with the federal government is 52 percent, and the actual amount collected is 24.2 percent. Meanwhile states are increasingly looking to research for purposes of economic development, but they are generally unwilling to pay much for it.

Now consider whether the revenue brought in for sponsored research helps fund the expensive science and engineering programs. In 2010, engineering at ASU generated $52,637,119 to cover the direct expenses of sponsored research. The indirect costs of this research were $33,161,385, at the rate ASU used in its negotiations with the federal government. Unfortunately, engineering brought in only $15,039,279 in indirect costs, for a deficit of $18,122,106.

In university lore, engineering would be credited for bringing in $67,676,397 (the total of the direct and indirect costs generated). Engineering did indeed perform and help fund an activity of crucial importance to the mission of the university and to the economy. However, the $18,122,106 differential was subsidized by the tuition paid by undergraduate students in the fields where costs are less than tuition. These subsidized costs should be part of the balance sheet when looking at engineering.

Engineering actually generates a higher percentage of its indirect costs than do other fields, such as education, where granting agencies and others are even further from funding full indirect costs. But if tuition and teaching load are factored in, the Teachers College at ASU, despite the indirect-cost issue, generates a profit. It lost about $2 million in indirect costs, but it made a profit of over $9 million on the teaching side.

Revenue: Fund Raising

Fund raising is often thought to be the great panacea. Annual giving, made up of many small contributions, provides a generally unrestricted revenue source that is spent immediately. Larger permanent gifts added to the endowment provide a longer-lasting revenue stream, but a gift to the endowment of $1 million, while generous and appreciated, generates only $50,000 in payout at a 5 percent rate, less than half the compensation of an assistant professor, including fringe benefits.

Large gifts do not come easily or often either, and frequently they require large expenditures by universities to obtain in the first place, a fact that is often ignored. The costs of the development office and officers in the university, colleges, and units should be presented along with the dollars raised. Gifts are also often in kind, not cash. What is really relevant is how much money is available at the end of the day.

And the end of the day may be a while in coming. Gifts sometimes come in the form of bequests that occur many years later. Even a $1 million cash gift to the endowment is usually not received in a lump sum but over years, say $250,000 a year for four years. Moreover, the $1 million gift is usually for a restricted purpose. Unrestricted gifts that can be used for the university's top priorities are very rare.

Nevertheless, in elite private institutions, annual giving and endowment funds function as an equivalent of state support in that they provide a stable revenue stream that can be even more important than tuition dollars. Because their endowments are so important, these institutions focus on individualized student attention, since successful and grateful students are sources of money in the future. Private institutions have small, selective student bodies partly for this reason. But they have to compete for these students and often discount tuition substantially to attract them (Leonhardt, 2008; Winston, 1997).

In state institutions, students are generally a much better and more reliable source of revenue than fund raising. So the focus is, or should be, on retaining students rather than on raising money from wealthy donors and cultivating future donors. At ASU's tuition rates, more money is earned by retaining 10 students for one year ($56,000) than is gained by a $1 million gift. And tuition is unrestricted; it is also recurring, if the institution's retention strategy is successful. The misconception that a $1 million gift immediately gives an institution $1 million to spend as it likes often leads administrators and boards to focus more on fund raising than on student success.

Comparing Universities

Comparisons are important for benchmarking purposes; it is hard to know if an institution is being as cost-effective as it might be without comparison to the performance of its peers. But different institutions have different missions and mixes of disciplines, so comparing them is not easy, nor is it always meaningful. A university with a medical school has different sponsored research opportunities than one without one.

Budgets vary by discipline, as we have seen, since both teaching and research are structured differently by discipline. So the appropriate comparison is between disciplines, not between universities. In the for-profit business world, this is akin to conglomerate corporations comparing their performance by benchmarking their individual business lines against performance in the sectors in which they each operate and compete.

There are also differences in how states operate that make comparisons across states difficult if not impossible. In New York, for example, the state pays fringe benefits and debt service directly for SUNY institutions, so those dollars never show up in the universities budgets; in Arizona each university pays these costs out of its operating budget. While these dollars show up in IPEDS reporting, Arizona's represent actual expenditures while New York's represent estimated amounts, since the state does not track these expenditures separately.

In addition, different universities apply different rules for answering the same questions on the same forms. For example, IPEDS collects faculty data on the Human Resources survey. In the past, it did so in three annual surveys: the IPEDS Instructional Staff/Salaries, IPEDS Employees by Assigned Position, and IPEDS Fall Staff. While a study of IPEDS Human Resource data quality (Clery, Arntz, Miller & Ratchford, 2008) found that the data were generally consistent with data reported elsewhere by the same institution, this does not mean that they are comparable between institutions.

For instance, for “instructional faculty” some institutions report all full-time ranked faculty, some include lecturers and some do not, some include only those paid 50 percent or more by state funds, and some define them as whoever is teaching (Gater & Lombardi, 2001). Some land-grant institutions consider their extension agents to be faculty members and report them as such, while others do not. The growth of branch campuses in foreign countries creates further reporting complications. IPEDS instructions say to exclude staff members who work at branch campuses, but an institution's budget structures and human resources systems may not make this possible.

What Should Be Done?

The academic side expends the vast majority of the money in higher education institutions (on average 70 percent in research universities, according to IPEDS). Without dividing the academic side into its units (colleges and departments), its functions (teaching, research, service), and its sources of revenues (state appropriations, tuition, contracts and grants, and gifts), no one can properly manage an institution or understand its true productivity.

In order to generate information that is helpful for analysis and action, revenues need to be divided by function and source, as do expenses (and all expenses must be included). Enrollment numbers, by level, should be included in analyzing instructional expenditures. The true amount of indirect costs needs to be factored into any analysis of research. And data need to be presented separately by college or discipline.

The National Science Foundation leads the way: It has started presenting its comparisons of research expenditures separately for schools with and without medical schools, since there is so much more funding available for medical than for other kinds of research. It also disaggregates data by the focus of the research. It recognizes that a university with a strong focus on physics and mathematics is not comparable to one focusing on biology.

With care, it is possible to get good data that is useful in managing higher education institutions. Decisions made with bad data are bad decisions. Only with good data is it possible to understand the true financial status of an institution, its performance, and its cost effectiveness.

Resources

1. Capaldi, E. D. (2009) Intellectual transformation and budgetary savings through academic reorganization. Change 41, pp. 19-27. (July/August)

2. Clery, S. B., Arntz, M. C., Miller, A. and Ratchford, S. (2008) Integrated Postsecondary Education Data System human resources data quality study, National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education., Washington, DC. (NCES 2008–150)

3. (2009) Expenditure analysis in state university system institutions, State University System of Florida Board of Governors.Retrieved from http://www.flbog.org/resources/iud/expenditure_search.php

4. Gater, D. and Lombardi, J. (2001) The use of IPEDS/AAUP faculty data in institutions' peer comparisons, Retrieved from http://mup.asu.eduThe Center Reports

5. Leonardt, D. (2008) The (yes) low cost of higher ed. New York Times Retrieved from http://www.nytimes.com/2008/04/20/education/edlife/essay.html?_r=14/21/2008

6. Romano, R. M. and Djajalaksana, Y. M. (2010) Using the community college to control college costs: How much cheaper is it?, Cornell Higher Learning Research Institute Working Paper 116, 2010

7. Winston, G. C. (1997) College costs: Subsidies, tuition and policy, The Williams Project on the Economics of Higher Education, DP-45, November 1997

Elizabeth (Betty) Capaldi is executive vice president and provost at Arizona State University. Previously, she served as provost at the University at Buffalo, SUNY, and then as vice chancellor and chief of staff of the SUNY system. Prior to that, she was provost at the University of Florida.

Craig Abbey is assistant vice president for planning and budget and director of institutional analysis at the University at Buffalo, SUNY. They work together in The Center for Measuring University Performance.

Read Comments  |   Submit Your Comment