Back to Dr. Johnstone's readings
I. Introduction
This paper is an introduction to basic concepts of economics as they relate to, and presumably are able to shed light on, the institutions and the issues of higher education. The intended readership is the practicing university administrator--president, dean, department chair, or admissions officer--or the serious student of higher education who wishes to better understand the workings of a university or some of the great public issues of higher education through the use of economics.
Neither this paper nor the lectures and discussions it has been written to supplement presuppose prior formal study of economics or accounting. Accordingly, some sections may seem elementary or even trivial to those familiar with the concepts, whether through formal training or wisdom gained on the job. To others, unfamiliar with such concepts as "present value" or "market equilibrium," the paper may seem an ambitious or even a presumptuous undertaking. Far more than a body of facts, concepts, or even analytical tools, economics is a perspective--a systematic way of thinking about issues, organizations, and various social phenomena. "Ways of thinking" do not come easily. Furthermore, the perspective that is the economist's is made more difficult to attain by the specialized language (of course, all disciplines have one), the quantitative language (a useful tool, but at times a cloak), and the enormous range of subject matters to which economics brings major insights and which are therefore occasionally mistaken for integral parts of the discipline.
But economics need not be all that mysterious, and certainly need not be intimidating. It is, indeed, an enormously systematic discipline. Its fundamental concepts are linked, often elegantly, one to another. But because the parts are so linked, the discipline reduces itself often to tautologies: seemingly striking conclusion simply arrived at once one accepts a few of the fundamental principles and a few rules for their application to the real world.
Economics is the study of resource use and choice: how a society, firm, institution, or individual under (a) constraints of limited resources, with (b) given (or assumed) tastes and preferences, and under (c) technological and organizational conditions prescribing the different mixes of output (guns, butter, higher education, etc.) possible from the available productive resources deploys those resources to the best ends--as measured by money, profits, happiness, satisfaction, student-credit hours produced, students graduated, or scholarly reputations.
II. Building Blocks
Before turning to a college balance sheet or a budget, or before entertaining a policy issue such as the appropriate level of tuition for a given school, let us develop five building blocks of economics as they pertain to the economics of higher education: costs, benefits, margins, time (or present value), and economic behavior.
Cost. In economics, we deal all the time with the costs of various goods, services, activities, and outcomes. But cost is not a simple concept. Cost is not always easily measured by our familiar unit of worth, "money." Cost is not always unambiguous. Cost is not the same as price.
The cost of something is measured by the resources given up to produce that something. But "resources given up" suggests hours of labor, minutes or hours of the life of a machine, parts of a truck or factory, and other such "apples and oranges," all of them hard to measure, and hard to compare, and hard to sum to a total. We can, of course, translate all of these resources into a common denominator of exchange, such as dollars, and thus express a total cost of a good or service.
But even this does not always give us a sense of what is lost by virtue of the production of the good or service in question. After all, what are dollars? what is money? The answer is simple: money is anything -- a piece of paper, a piece of precious metal, or "credit on account" -- that can claim goods and services produced, and offered for sale, by others. The cost of a $10 good is, to be sure, ten dollars. But the cost may also be viewed as the loss of whatever else could have been secured with that $10 claim. In many ways, the best measure of the cost of X is the lost benefit to us of not having Y, Z, or W (the best of these) that we could have produced, or bought, with the resources, or money, devoted instead to the production or the purchase of X. In other words, the cost--the opportunity cost -- of X is the lost benefit of the best alternative now foregone because of the devotion of productive resources to X rather than to something else.
One of the costs of higher education, for example, is the loss of production by students who, were they not in higher education, could presumably be out in the real world contributing at this very moment to the Gross National Product. What is foregone by their being students, then, aside from all the goods and services that could be produced by the labor of the faculty and administration, the physical plant of the college, and all of the resources poured into the production of books and materials, is the potential product of the student labor, best measured by their foregone earnings.
Let us look at opportunity cost in higher education from another perspective. The cost of a new freshman seminar program may be expressed in the dollars necessary to buy some books or to pay for the time of the teaching assistant, faculty member, or administrative coordinator. Another, and perhaps more useful, measure of the cost of this new program is its opportunity cost -- the loss of all the benefits of the best alternative program that could be run with the faculty, staff, and cash about to be devoted to the proposed freshman seminar program. Is the seminar program worth foregoing the benefits of this next best alternative? To say "no" is just another way of saying that the benefit of a seminar program is not worth the cost -- that is, is not greater than the benefit of the best alternative. To assert that the program is worth the cost is simply to say that its worth or benefit is at least as great as the worth or benefit of the most worthwhile or beneficial program or use that could alternatively be made with the program's budget, or cost.
Even the most cursory treatment of cost in higher education is incomplete without a mention of full cost, or the sum of the direct plus indirect costs, of a particular program. The direct costs, for example, of our freshman seminar program, are easy to see and to measure. They would include faculty salaries, special materials, and possibly the part-time services of an administrative coordinator. However, there are very real, but much less visible and measurable, costs, some of which can also be attributed to this new program. These are costs of the heating and lighting of the building, debt service on the plant or on the school's accumulated deficit, maintenance costs for janitorial and other services, and the like. Or, we can enter into the realm of even more distant, but nonetheless real, indirect costs: the cost of the blanket insurance policy for the entire university, the cost of the offices of the president, provost, controller, treasurer, and personnel director. These and a wide range of other indirect costs cannot easily be linked to a particular program much less to a particular level of activity of a particular program. However, they are incurred in the end only for the support of educational programs such as our freshman seminar, and we ought in some way to be able to distribute all of the indirect support costs to the true "end use" programs for which the college or university operates. (The reader may quickly realize that a new program does not necessarily create new indirect cost--a perception to which we will return in the section below on "margins.")
We must not leave this section on "cost" without a reminder that "cost" is not the same as price. Cost is what is actually given up in the production of something. Price is what a user has to pay for it. Normally, a producer prices his good or service to recover the full cost (including a profit which, for now, we will assume to be part of the cost). But prices may frequently depart from costs -- for example, when the producer is a public agency whose costs are partly subsidized by the taxpayer and only partly (if at all) met by prices paid by users--or when certain costs of production such as pollution of the air and water do not have to be met by the producer and do not therefore have to be reflected in the price--or when prices are set high enough to cover special taxes or "excess" profits unrelated to the cost of producing the good or service. In such cases, productive resources generally wind up being allocated in a way other than that which would have been had prices and costs been equal. The result may be for good or ill, and that judgment may depend on the priorities of the judge. But now we are at the heart of what may be called public finance, and it is time to return to our economic building blocks.
Benefits. Benefits are what we allocate resources for. Benefits are why we produce. Indeed, without benefits, there are no costs, since the cost of one enterprise is ultimately the inability to reap the benefits of another foregone by the choice of the first.
Benefits can be conceptualized however the reader would like. They may be monetary, which means simply claims, in the form of coins, currency, or checking account balances, on real (i.e., consumable) benefits to be realized at some other time or place. Or, they may be non-monetary, which may include such worthy attributes as happiness, satisfaction, pride, or honor. Do not be misled by those who like to insist that honor, for example, has no price. It almost always does, although that price may be very, very high. As we engage in activities that bring increasingly more satisfaction, pride, or honor, we may have to give up so much (of other benefits, monetary or non-monetary) that the choice is made to forego at least that additional pride, honor, or satisfaction. There may be, at least in the best of us, some pride or honor that has no price -- that we would not exchange for any amount of any other satisfaction. But most assuredly these virtues have a price on the margin -- that is, that are applicable to the last unit added -- and most assuredly at some point the worth of these additional benefits begins to decline and one begins to entertain at least the possibility of a trade-off with some other benefit, whether yet directly measured in money or in some other less easily measured form of satisfaction.
Benefits may be realized immediately, or may be anticipated now for some future point in time. Almost always, the benefit now of some future goodness is less than the benefit now of some immediate goodness: in short, a bird in the hand, and all that. Or, as we shall see below in the section on Time, future benefits get discounted back to the present. But more on that in a moment.
To speak of present or future benefits raises the concepts of consumption and investment. Investment, to an economist, is nothing more than a devotion of resources not to the production of a good or service for immediate use, but rather for the production of a new productive resource, then to be employed for the production either of a consumable good or service, or for yet another productive resource, and so forth and so on. The point of all this is that consumption, in ultimate sense, is the end of all production. Yet man quickly learned years ago that by foregoing some immediate consumption and by devoting those resources instead to the production of machinery, roads, dams, irrigation projects--or, yes, education and training -- one could so enhance productive capability as to reap in the future not merely an output of consumption goods equal to those which were deferred at the time of the investment, but with a return which, after all, is only reasonable if one is to undergo the inconvenience of deferring satisfaction and the risk of not recouping the lost, or foregone, benefits at all.
Benefits may be private, when they accrue quite directly and quite fully to users, who generally, but not always, have explicitly called and paid for the good or service in question. Benefits are public when at least some benefits to a particular activity accrue to those not easily identifiable as users and hence not easily assessed or charged for the benefits received. Most food, clothing, and shelter, are private benefits, easily linked to the user or consumer, and just as easily and fully charged to them. Public safety, weather reports, national defense, and city streets tend to be quite public--that is, enjoyed by a broad and virtually unspecifiable range of users, making the selling of the goods or service by a price to specific consumers quite impossible. Private benefits are associated with the production of goods and services that are on the market, sold directly to the consumer at prices generally reflecting the costs of production. These goods and services can be thought to be "worth the price" by the simple observation that people would not be paying that price were not the particular good or service thought by the consumer to be at least as beneficial as the next best alternative obtainable for the same price.
It is considerably more difficult to assert unequivocally the worthwhileness of goods or services that are primarily public benefits and that have not been purchased as such by users and that have not stood the acid test of a direct comparison with all other goods and services that could have been bought instead. Public goods and services are purchased by elected and appointed officials on behalf of "the people." Yet, are they perceiving the proper benefit? And are they comparing the benefit to the benefit of the next best thing -- including, presumably, a tax cut?
The extension to higher education is obvious, although conclusions are anything but simple. Higher education is a consumption good with many private benefits. Higher education is also an investment good, again with much, although not all, of the future benefit capturable by the private user, or purchaser, of the higher education. One of the major public policy issues today is the degree to which higher education is also a significant public good; that is, a good the benefit of which does not accrue simply to the students, but accrues also to a population beyond those students such that the non-student population (call them "the general taxpayer") can appropriately be taxed to subsidize the purchase of education for the direct (but only partial) benefit of the student population.
Margins. The economist spends much of his or her time at the margin. Rather, all of us spend much of our time at the margin, weighing not the costs and benefits of all the housing, all the education, all the clothing we might buy, but rather the benefits of an extra unit, or the cost to us of losing just one of the units that we are now enjoying. When we talk about extra or last, we are talking on the margin. The significance of marginal analysis is that both the costs and benefits of a given unit of a good or service, whether to an individual or to some collectivity of individuals such as all consumers, generally vary--that is, increase or decrease -- as the number of units is increased or decreased. Thus the cost or benefit of the last, or marginal, unit is likely to differ from the average cost or benefit of a unit calculated on the basis of all units consumed or produced.
A particular unit of output -- call it a student credit hour's worth of teaching--is composed of costs some of which are variable, that is, which increase or decrease roughly in proportion to the units being produced (e.g., the cost of faculty), others of which are fixed, that is, which remain relatively unchanged over large numbers of units (e.g., indirect administrative costs). The greater the proportion of costs that are fixed, the more those costs can be spread over large numbers of units, and the more the marginal cost will drop below the average cost. This is simply the common phenomenon of increasing returns to scale. In contrast, where costs are mainly variable, marginal costs may differ little from average costs, and there may in turn be little return to scale or size alone.
Time. Cost may be incurred and benefits received over periods of time. In order to weigh cost and benefits, we must make these time periods comparable. A $40,000 machine that brings in $8,000 in new revenue during the first year will seem like a bad proposition until one realizes that the machine will probably also bring in $8,000 a year after that, when it costs nothing, and quite likely the same the year after that for some years until the machine wears out, making it seem then quite profitable. Obviously, the trick to assessing the costs and returns of this machine is to distribute the cost over the years during which the returns will also be made. For example, if we borrow money to buy the machine, we could amortize the principal and interest over the years of useful life of the machine. If the machine could bring in $8,000 a year for 10 years and we could amortize $40,000 a year at whatever interest rate we had to pay for $8,000 a year or less, the benefits would obviously be worth the cost. Similarly, if we had saved for years a pot of money with which to buy machinery, we could sink that pot of money into our new machine. But would it be profitable? Since we are not literally paying for the machine over time, our test will look a little different than the example just mentioned. But only a little! For if we are prudent, we had better begin immediately putting enough money aside to build up a pot of similar size to draw on when our new machine wears out. We might call this pot our depreciation fund. Notice that putting money aside to depreciate our machine is exactly the same as amortizing a loan or making payments on time. The fact that in one case we borrowed and in the other case we used our own funds has no bearing upon the way in which we account for the costs and benefits. In the latter case, we are simply borrowing from ourselves.
However, we cannot always keep costs and revenues or benefits so nicely in line. For example, a year's worth of revenue from tuition and other sources supposedly is equal to a year's worth of operating costs. However, costs are incurred relatively uniformly throughout the year, but revenues come in large chunks when tuition bills are paid, state appropriations credited, and gift pledges actually received. Even within the framework of a single year, timing can make an extraordinary difference if we must, for example, go to the bank to borrow for our payroll in July and August and if, in September and January, we have large bank balances drawing interest from the tuitions just paid. In either case, we need a convenient way of measuring a flow of costs and a flow of revenues, and some way of controlling for the differences in the timing of these flows.
The device the economist uses is the concept of present value. The present value to me of $100 a year hence is the value to me now of that $100 that I will not actually get until a year from now. Put another way, it is the amount of money currently that I would be willing to exchange for a promise to be paid $100 one year from now. It is reasonable to assume that I would not willingly give up $100 now for the promise of $100 a year from now. At the very least, I could take the $100 that I have now, put it in the bank to earn a year's worth of interest, and find that I had something worth more than $100 twelve months later. But there is certainly some amount of money--albeit less than $100--that I would willingly give up now in return for that promise to pay a year hence. And that amount of money is the present value of a $100.
That future money is worth less in the present is obvious, yet it is instructive to untangle the reasons why. One reason is risk. The more risk there is attached to the prospect of that future money, the less one is likely to be willing to give up now for it--i.e., the lower the present value, or the more the future value is discounted to the present. A second reason is inflation. The more the dollar is anticipated to undergo the erosion of inflation, the less willing one is to defer use of the money at present in return for a promise to pay in the future; again, the more that future sum is discounted, the lower is the present value of the future dollar. Finally, and the most basic reason of all, money has value. Even in the absence of risk, and in the absence of any inflation, money has a value as long as it can be devoted to the production of some kind of capital good that will, in turn, yield greater output in the future. Thus, money has value and consequently will have an interest rate above zero even in the total absence of risk or inflation. Such an interest rate, minus that which is necessary to compensate for inflation and risk, would be a rather modest one, equal to the assumed real (that is, after inflation) rate of growth of the economy.
In fact, all of these elements --risk, prospective inflation, and real rate of return on capital--are reflected in the prevailing rate of interest. Thus, the prevailing interest rate (ignoring for the moment the fact that there are several) is rate by which, for most purposes, we can discount a future sum of money back to the present for the purpose of comparing it with other sums of future money similarly discounted back to a single and comparable point in time--namely, the present.
Behavior. Economics is above all a behavioral science. That is to say, it is about the way people, through their individual and collective behaviors, affect the use of resources.
Although this assertion is so obvious as to seem trivial, economics is replete with the anthropomorphic explanation. We hear all the time that prices rise, costs go down, production slows down, and demand picks up when we ought to know perfectly well that prices, cost of production, and demand, do not "do" anything whatsoever. Rather, people raise and lower prices, charge higher or lower costs, decide how much to produce, and offer to buy more of a good at a given price thereby increasing demand. All of the things that we assume to happen by our economic precepts do so because we are assuming certain kinds of behaviors on the part of people. For example, everyone who has taken a course in principles of economics (and most who have not) knows that prices go up when demand exceeds supply. But we just learned that the prices do not "go up" at all; rather, some people raise prices and other people continue to buy at the higher prices. We must ask, then, what the behavioral assumptions are which allow us to predict such a result. And they are very simple. The "increase in demand" is, by definition, an assertion that more people are willing to pay higher prices to purchase a particular good. The assertion that prices will be raised follows upon the assumption of profit -- maximizing behavior on the part of the producers, whom we assume will prefer to raise prices and make more money as long as people are willing to pay them rather than to not raise prices. And we assume that raising of prices will continue until that point at which there are no more unsatisfied buyers waiting outside the stores with unmet demand and money in their hands.
What a simple model! What an obvious outcome from these simple examples. Yet, what an extraordinary and complex set of assumptions underlies that simple exercise: e.g., that buyers make their own demands known to producers (or perhaps, in this case, to distributors)...that producers maximize profits rather than good will, stability, market shares, or some other things that could, with some plausibility, be the drummers to which they march...that producers know how much consumers will buy at what price...and that new production is possible in response to the increased demand, unconstrained by limited supply, by prohibitive technology or front-end capital needs, or by other constraints upon production.
Now let us look at the behavior which drives the decision makers in an organization that does not run on the simple profit-maximizing incentives of our elementary market equilibrium model. How are resource allocation decisions made in a hospital, a government agency, within the military, and in universities? Who are the people who make the decision that allocates resources? To what do they respond? What are their rewards?
Look, for example, at a hospital. One of the major incentives within a hospital may be the minimization of "therapeutic misadventures" -- i.e., the avoidance of mistakes that cost patients' lives and cost the hospital heavy lawsuits. To behave in such a way is probably to incur very large costs for multiple tests, expensive equipment, and round-the-clock care of a kind that might be considered expensive -- that is, not worth the benefits foregone -- in any system other than a hospital, which recovers such costs from insurers and other third-party payers and which may not be otherwise rewarded for efficiency or cost reduction. (Such behavior goes part way to explaining the Diagnostic Related Group [DRG] method of third-party payment for medical procedures.)
Another instructive example of behavior in a public institution is the manpower training center, the administrators of which may be rewarded according to the percentage of successful job placements from its participating class. With such an incentive, those running the agency are, with the best of intentions, likely to try to enroll "winners" -- that is, already well-motivated and intelligent trainees -- in order to maximize their successful placement, and to avoid those high-risk individuals whose training in fact might be the most important, yet whose successful but only occasional placement is simply not rewarded in a way commensurate with the avowed goals of the agency.
The military is replete with classic examples of incentives working at cross purposes. In World War II, the European field commanders were rewarded ultimately for battlefield success, and looked upon the transport of troops as a way to put the greatest number of front line and support men at their disposal on the Continent. The Navy, while obviously under the same incentives to win the war, was largely rewarded, all else being equal, by minimizing the loss of its men and ships. The Joint Chiefs and the civilian leadership were rewarded by, among other things, keeping down casualties and other catastrophes. Obviously, the way large numbers of men were moved to Europe through enemy submarine-infested waters could vary considerably depending upon which set of incentives most affected an operation.
Let us look now at higher education. What are the operative incentives affecting the behavior of faculty, department chairmen, and deans of a university? What they generally, and understandably, want are the following:
What are generally not rewarded, at least at the departmental or even at the school, level (setting aside, for the moment, the profound question of the degree to which good teaching is rewarded) are such achievements as:
While the latter behaviors are not necessarily virtuous nor the former ones profligate, it is obvious that an incentive system that either ignores costs or works generally toward their increase can only over time come into serious conflict with theconstraints of limited resources. The conflict is particularly pronounced in a college or university, where so much decision making is in the hands of faculty who do not have responsibility for the financial health of the corporate institution.
One approach to this dilemma is to accept a division of responsibilities: academic content for the faculty, and financial viability of the institution for the administration -- and to hope for the best from the resulting creative tension. A different approach would be to build incentives for the decentralized decision makers (namely the faculty) to economize on cost and to achieve institutional as well as departmental and personal ends. For example, schools and departments can be given a firm budget and made to live -- and to maximize its own benefits -- within it. Another is to build budget incentives around institutional objectives -- e.g., student retention, or the acceptance of transfer or university or other "target" category of student. For example, campus or school or departmental budgets may be driven by enrollments, but those enrollments can be weighted in various ways to enhance the attractiveness, to the particular unit, of attracting and retaining certain kinds of students.
III. University Finance: Sources of Revenue and Categories of Expense
Let us now look at the finances of a college or university, drawing on these and other economic concepts. The beginning is simple. Over a period of time, the school spends income, which it gets from a variety of sources, for its various purposes or programs. To understand the finances of a college or university we must understand these components of income and expenditures, how they relate to one another, and what their future rates of change are likely to be.
The sources of income and categories of expenses are best seen in the institution's operating statement. This is a statement which, in retrospect, itemizes the revenues accounted for in the time period (usually a fiscal year) with the objects or categories of expenditure over the same time period. This is the accounting statement closest to a budget, that more familiar document that we might describe as a planned, or prospective, operating statement.
Sponsored Programs. An easy income and expense category is sponsored programs. These are funds paid to an institution for services rendered: for the deployment of faculty time, and the expenditure of other funds for, e.g., graduate students, equipment, clerical services, and the like for specific research or training projects. There are, in addition, the indirect costs associated with the sponsored project: costs which may not easily be directly assignable to a particular research project, but without which the capacity to perform the research would, in the long run, not exist or not exist in the appropriate quality. Such costs include utilities, administrative services, building maintenance, and the like.
The calculation of an appropriate indirect cost rate is both science and art, but must, in the end, be explainable and auditable. For example, if personnel paid for by direct sponsored research budgets constitute one-third of all university personnel, it would seem appropriate to charge one-third of the costs of the central personnel officer and the payroll department to "research." Similarly, if 40 percent of the number of accounting transactions involve sponsored research budgets, that percentage of the direct costs of the comptroller's function could appropriately be assigned as an indirect cost to research. Usually, such rules-of-thumb are devised, and costs then tallied and audited for a complete fiscal year. Then, the allowable costs are divided into direct salary budget to give a ratio that, at least for the year under audit, reflected the proper ratio between the easily and directly measurable direct costs and the supporting indirect costs. For the coming year, the federal government generally agrees to pay a similar percentage of direct costs in anticipation of appropriate indirect costs. University research administration offices have no end of difficulty explaining to faculty members why a certain percentage of faculty salaries on their specific grants must be set aside for indirect costs, even when a particular project may in fact draw on none (the problem of "marginal" thinking), but the overall costs are real, the appropriateness of attaching them to sponsored research is unquestioned, and the formula method of setting aside a percentage of salaries to assure their recovery is a generally acceptable and convenient short cut.
Auxiliary Enterprises. Auxiliary enterprises are similar to sponsored research expenditures in their direct link between the income source and the expenditure. Normally, auxiliary enterprises such as the dining halls, the residences, the bookstore, and the parking lots are expected to pay their own way, including coverage of indirect costs calculated in ways similar to the methods used for assigning such costs to sponsored research projects. But let us not accept a rule of "pay your own way" simply because it sounds virtuous and responsible. Why, indeed, should auxiliary enterprises be sold to users at a price which equals their cost. Why should they not be sold at a lower price, with the difference coming in the form of subsidy from other sources of revenues? Or, why should they not be sold at a considerably higher price, bringing profits which could then be devoted to other programs?
The answer was anticipated in our sections on cost, benefits, and incentives. People behave -- in this case consume resources -- according to price not cost. The lower the price the more (everything else being equal) will be consumed. Thus a subsidy, whether of a dining hall or a residence operation, will generally lead to the consumption of more of the good in question than would otherwise be the case. The extra costs, of course, are incurred anyway, even if not recognized by the direct consumers. And the costs, as we described earlier in this paper, are most usefully viewed as the loss of all of the benefits that could have been produced and consumed with the best alternative use of those resources that went to the "extra" production of the subsidized good or service.
Consider the case of the dining halls. Without question, the dining halls could be priced below cost, more and better food and services produced (and happily consumed), and the losses recovered from tuition, state funds, or other sources. It is not absolutely clear that there would be any inefficiency or, indeed, any other loss as the result of such an arrangement. For example, a college may require all students to eat in the dining halls, requiring dining contracts as well as tuition from all students, and therefore making little real distinction between the tuition and the dining charge on the students' bills. However, in the case of a dining service that is not mandated nor universally subscribed to, a significant subsidy raises two problems.
The first is a possible loss of efficiency - that is, the possibility that more resources are being devoted to the production of dining than the benefits justify. If there is a subsidy, as revealed by a consistent operating loss when all costs and revenues are appropriately assigned to the dining service, we ought to begin wondering what would happen were price to be increased to equal cost? Some students might stay away, those being students for whom the benefits of college dining were not that great in the first place, and for whom the added price tips the scales in the favor of alternative forms of sustenance. Who, we might wonder, are we to say that this is wrong? (One answer, of course, and not a trivial one, is that students may not in fact be competent to provide nutritionally balanced meals on their own, and that the college has an obligation to encourage good diet by the inducement of a subsidized meal.) Or, students might stay away, leaving the dining service to begin losing money, which in turn might lead them to either change the mode of operation, offer different menus, reduce the amount of food served, or make some other appropriate step to regain lost customers. Again, why would we want to forego the kind of signal that might allow all this to happen? In other words, an internal market system can be a potentially valuable signal for the alignment of costs and benefits. A significant departure of price from cost via a subsidy (or an excessive profit) can easily lead to an allocation of resources considerably different from that which would prevail in the absence of the subsidy or excess profit. If this departure is in line with institutional aim, the subsidy or profit is, of course, a good thing and a valuable institutional tool. If it is not, then we may, by not running our operations on a break-even basis with price equal to cost, be deploying our (and our students') resources less than optimally.
In addition, as in so many similar examples, a question of equity is raised alongside the question of efficiency. Someone is obviously paying the cost of the subsidy being used to reduce the price of dining service below its full cost. Presumably, this "someone" is either the non-dining service student, the general taxpayer, or some other non-beneficiary of the service. Most of these people (particularly the general taxpayer) are probably less able to pay than the student, and although dining services everywhere have their detractors, it is not clear that there is any particular social equity in having the apartment-dwelling student or the general taxpayer contribute to the support of those students eating comfortably in the college dining hall.
IV. Education and General Expense
The core expenses of a university are those faculty and support salaries that are not covered by sponsored research grants, plus most administrative and other supporting costs, including, for some private institutions, debt service and depreciation. These expenditures may be further broken down into such categories as: instruction and departmental research (primarily faculty salaries), library, academic computing, student services, general administration, and operations and maintenance.
Tuition and Fees. Income to cover these core operating expenses comes from four principal sources, the relative importance of each depending upon the type of institution: public or private, college or university, well or poorly-endowed. The first source is tuition and fees, which cover anywhere from 65-85 percent of the education and general costs in a private college and from 15-30 percent in a public four-year institution. Income from tuition, in turn, is a function of three variables: (a) the catalog, or stated, tuition, multiplied, times (b) the enrollment, less (c) that portion of otherwise unrestricted revenue, presumably from tuition, that is turned back to the students in the form of financial aid. (The return of tuition income to financial aid is rare in public institutions.)
The calculation of net tuition revenue, or tuition revenue less that amount discounted in the form of grant aid, gives us an opportunity to use some of the concepts treated earlier in this paper. For example, increasing amounts of financial aid may be brought in by the aided student, perhaps holding a portable state grant or a federal Pell Grant, or perhaps from institutional sources legally restricted to student aid. Such aid does not come from the unrestricted revenues of the university or college, and is therefore not properly netted from gross tuition revenues. In fact, such aid can be said to have no opportunity costs to the university in that there was no alternative use of those resources and therefore nothing foregone by virtue of the grants provided. One might still, however, talk about the opportunity cost of a student aid fund restricted to students from New Jersey: the cost of granting any available funds to one particular student could be viewed as a denial of those funds to some other eligible student -- providing, of course, there is another worthy New Jerseyan in the applicant class. And there is, of course, a significant opportunity cost to society, since the resources claimed by the state or federal grant could ultimately have been devoted to some other purpose. (This, fortunately, need not be the concern of the college or university planner.)
But even financial aid which comes as a direct rebate from tuition revenues can be shown, in certain circumstances, to have little or no opportunity cost. For example, fluctuations in student interest in certain fields (e.g., undergraduate engineering in the early 1970s, or teacher education in the late '70s and early '80s) can lead to such temporary low enrollments that the enrollment of an additional student arguably adds no additional costs whatsoever. The marginal costs in such an instance would be zero, or even negative. Since the marginal return to a financial aid program that directly increases enrollment may be calculated as the catalog tuition less any money rebated as financial aid, the marginal revenue is positive so long as the grant does not exceed tuition. In other words, and in very simple and even trivial language, any net tuition received at all from students in such circumstances may be viewed as net revenue to the institution.
Government, or Taxpayer. The second principal source of funds to cover educational costs is the government, more appropriately designated as "the taxpayer." Federal funds to higher education are virtually all devoted either to student aid (much of which, of course, enters the general income stream via tuition), sponsored research, or special restrictive programs. The bulk of governmental or taxpayer support for educational expenses comes through state appropriations to public institutions, based primarily upon enrollments, generally differentiated by level and programs.
Endowment. The third source of funds, observed mainly in a relatively small number of private institutions, is return on endowment and other investments. The term "endowment" refers generally to funds held in trust by the institution (specifically by the trustees of the institution) in perpetuity, the income, interest, dividends or yield from which are to be used for current operations, either restricted or unrestricted, depending upon the terms placed upon the endowment by the donor.
Many institutions are fortunate enough also to have certain funds on hand in the form of reserves. These funds, like true endowments, are invested generally with the aim of maintaining the principal amount intact, but unlike true endowments, there are no legal obligations to preserve the principal.
Such funds are often designated quasi endowments, or funds functioning as endowments, and may be liquidated by the trustees should the occasion warrant. However, no institution can survive without some legally unencumbered reserves. It would also be imprudent to liquidate reserves for normal current operations without an intention to replace the reserves as soon as possible; whatever caused the need to tap reserves in one year may well occur the following year, and in fact may be said to be more likely to occur since there is now less income due to the depleted principal amount of the funds functioning as endowment. Therefore, while a shortfall of income from other sources or an unusual increase in certain costs may require the trustees of the private college or university to draw at times upon the principal amount of funds functioning as endowment, a prudent board will usually insist that this transaction be considered an internal loan, to be repaid with interest in future years in order to assure the maintenance of the necessary reserve fund.
Annual Giving. A fourth source of income for the support of core educational costs is annual giving for current operations. This source may be particularly volatile in small private colleges that depend heavily on gifts from a relatively few individuals, whose year-to-year fluctuations in giving (perhaps occasioned by stock prices at the end of the tax year) may cause considerable fluctuations in this income source.
Gifts for current operations are often restricted, just as are endowments--and, indeed, as are funds for sponsored research or other programs for which funds are designed by the donor. Other things being equal, a simple rule of thumb is that a college or university administrator would prefer unrestricted to restricted funds. At the same time, the restrictions on the use of funds may not always have a real impact -- at least not always the impact intended by the donor. For example, a gift of either endowment or operating funds to a private college, restricted to undergraduate student aid, may well do little more than reduce the draw that year on unrestricted funds for financial aid. The actual effect of such a grant, seemingly restricted to undergraduate financial aid, may thus not, in fact, be to increase the amount of aid available to undergraduates, but to do that which is now made possible by the release of otherwise unrestricted funds formerly obligated to the financial aid budget.
Borrowing. A fifth source of income for the support of university operations is borrowing. When costs are incurred--that is, checks actually written and the university's account drawn down -- in amounts that exceed the revenue brought in, some funds must have been borrowed to make up for the shortfall in income. These funds may have been borrowed externally, from a bank, or may be borrowed internally, from the institution's own endowment or quasi-endowment funds. In order to borrow prudently from one's own endowment, the loan must be as financially sound as any other investment that might have been made with the endowed funds--that is, at a proper rate of interest, with assurance of repayment and perhaps even with security. Borrowing from quasi-endowment imposes no such legal obligations for repayment; indeed, quasi-endowment can simply be liquidated to cover the year's operating deficit. To treat the transaction as a loan, however, signifies a determination to replace those reserves as the institution corrects whatever condition led to the precipitating gap between planned income and planned expense. Whether to borrow internally or externally is an investment portfolio decision: if the cost of external borrowing is reasonably low and the present portfolio of stocks and bonds a good one, the college might decide to keep its funds where they are and borrow from a bank. If bank rates are high and the endowment and quasi-endowment relatively liquid (that is, easily converted to cash), internal borrowing may be the better use of funds.
V. The Costs of Higher Education
Costs of higher education have increased greatly in virtually all of the post World War II years save the decade of the 1970s. Much of this increase in cost can be explained by a very great expansion in output, measured in any number of ways -- for example, by the sheer number of students, the advanced level of education and degrees granted, the quality of education and research, the social value of the education of new kinds of students, and so forth. But given any stable measure of output -- such as per credit hour or per full-time equivalent student enrollment -- and keeping all the factors constant, there has also been a very great increase in unit costs.
It is the increase in the unit costs, which most concerns those who spend much of their waking hours either studying or worrying about the financial health of higher education. Five principal factors, not all of them mutually exclusive, lie behind this increase in unit costs, again controlling for differences in the quality of the product or in the costs of educating various kinds of students or for various kinds of degrees or at various levels of postsecondary education.
The first of these factors we may call inflation generally. Whether we take the Consumer Price Index, which is a measure of the increase in prices of the average "market basket" bought by the average middle-class urban family, or any other general measure of price increases, we have become used to costs and prices going up, and it is obvious that higher education, like all households and institutions, will be affected.
The second reason is what we may call inflation particularly. The conventional measures of inflation, after all, are averages of "market baskets" of goods. But the market basket of goods that a college or university buys is a very special one. During most recent years, the prices of the collegiate market basket rose at a rate considerably faster than the market basket of the urban family of four that defines the familiar Consumer Price Index. In part, this is simply because so many of the goods and services that colleges and universities buy were themselves leaders in the inflationary band: fuel (especially for northern residential schools), books and periodicals, paper, scientific equipment, interest, and insurance. Since any inflationary index is an average of many price increases, and since an average increase implies some increases that are above and some that are below this mark, it should not be too surprising to find a particular sector -- in this case, higher education -- that tends to need goods and services from the "above average" side of the CPI and that itself experiences a rate of increase of costs greater than inflation generally.
But there is a third and an even more pervasive factor behind tendency of unit costs in higher education to outrun unit costs and prices generally. This is the phenomenon of rising relative unit costs in the labor-intensive sectors of the economy.
If we assume that wages throughout the economy rise at approximately the same rate regardless of the increase in productivity within any particular sector, and if we assume that the unit prices rise at the same rate as unit costs, we may state the following two propositions in the form of equations:
1. (W = (P + (O
and
2. (C = (P = (W - (O
where
3. (W = average change in wages, economy-wide;
4. (P = average change in prices, economy-wide;
5. (O = average change in productivity, or output per worker, economy-wide; and
6. (C = average change in unit costs, economy-wide.
In other words, the change in average unit costs is equal to the change in average prices which, in turn, is equal to the change in average wages minus the change in output per worker.
However, in the productivity immune sector, there is, by definition, no measurable increase in output per worker. Thus, the increase in unit costs in this sector (which includes higher education) equals the increases in average wages--with no offsetting productivity effect. To return again to equation form:
7. (C (Higher Education) = (W (Higher Education)
which equates the change in unit costs in higher education to the change in wages in higher education. Assuming wage increase in higher education equal to wage increases economy-wide, and substituting equation "1." for the right-hand side of equation "3."
gives us:
8. (C (Higher Education) = (P + ((O
In other words, the rate of increase of unit costs in higher education is equal to the rate of increase in prices generally, or the rate of inflation as measured by, say, the Consumer Price Index or the GNP Price Deflator, plus the rate of increase of output per worker, or the real growth in the economy. Since the real annual growth of the economy has averaged approximately 2-1/2 percent over the greater part of this century, we may say that a natural rate of increase of unit costs in higher education is equal to the prevailing rate of inflation plus 2-1/2 percentage points, or whatever might be the then-prevailing rate of increase of output for the economy as a whole.
However, unit costs in higher education in the period of the mid-sixties to the early 1970s and again during much of the 1980s rose not only at 2-1/2 percentage points, but at anywhere from 4-8 percentage points above the rate of inflation. Our equations above give us some clues for why or how this must have happened. In fact, wages in higher education during these periods did not simply rise at a rate equal to the rate increase of wages generally but at a considerably higher rate, which led to an overall improvement in the compensation of the professorate relative to other professions. This above was sufficient to push the rate of increase of costs in higher education past the "normal" excess of 2-3 percentage points above the rate of inflation. In addition, not only was there no productivity in higher education, but "productivity" by conventional measures such as course credits for full-time student enrollment per faculty member and administrator actually declined in this period--meaning that more services or more capital or more faculty time was made available per student. Here, too, forces at work to push the rate increase of unit cost in higher education up even faster than their normal 2.5+ percentage points.
By contrast, during much of the 1970s--years of austerity and belt tightening certainly unprecedented in the years since the end of the second World War--the rate of increase of unit costs in higher education dropped to a point where it was virtually identical to the rate of increase of cost and prices generally. In other words, the normal "plus 2.5 percentage points" disappeared. But our equations above tell us that this can only happen with a lower than average salary increase in higher education and/or with some productivity increases in what is thought to be a productivity immune sector. In fact, the explanation for much lower rate of increase of unit costs in higher education during that period lies with the former. Wages and salaries in colleges and universities, both public and private, rose in the 1970s at rates considerably below the rates of increase of wages and salaries overall. And without question in some sectors, student/faculty ratios have risen, suggesting (again by our crude measure) an increase in productivity in the higher education sector. Costs were stabilized in education, quite simply, mainly at the expense of the relative compensation of the professorate. And the explosion of unit costs in the 1980s was mainly a reflection, at least in some institutions, of "catch-up" wages and salaries, new investments in equipment and capital, and new administrative and student services.
Technology and Unit Costs. Productivity increases in the larger, economy are fueled substantially by technology. At an earlier time, this might have meant machines and industrial processes that magnified the output of the worker. Nowadays, it is as likely to mean computers, storing and disseminating information, and replacing, not simply magnifying, the output of men and women. Clearly, technology has a large and growing place in higher education: from what is taught, to how and from where researchers find and analyze information, to the storage and retrieval of personnel and student information, to how the buildings are heated, cooled, and kept secure. But can--or will--technology reduce unit costs?
The record of technology in reducing unit costs is not good. For the most part, technology allows us to do better, or to do whatever it is that we do more easily or more conveniently--which is, after all, a clear increase in productivity. But it rarely does it more cheaply. Word processing, for example, is incredibly more productive than typewriters. But the consequence of the phenomenal increase in word processing capacity brought about by personal computers and word processing software is that we write and print vastly more, and the products looks immeasurably better. An increase in productivity, to be sure. But our u it costs have not gone down.
Many individuals look to instructional technology to reduce the unit cost of instruction. In theory, there is unquestionable potential. the personal computer with new, sophisticated, and mass produceable software, ought to allow more learning to proceed self-paced, in the learner's home, at the learner's speed, rather than in classes, at rigid times, with live (and paid) professors Similarly, distance learning technology, such as multi-way, fully interactive video, can, at least in theory, allow professors to reach learners at distance sites, allow greater institutional specialization and division of labor, and bring down the unit costs, especially of advanced classes taught traditionally in small redundant groups.
However, most of our experience with instructional technology shows that it, too, like word processing, provides enriched instruction, or perhaps more convenient learning, but rarely cheaper instruction. This, of course, is also an advance in productivity: an increase in learning for the same, or maybe only slightly greater, costs of instruction. And it may well be that lower unit costs as a result of technology will only come about at a scale, and over a period of time, that we have yet to see implemented and adequately measured. But instructional technology has rarely been demonstrated to bring instructional unit costs down in the short run or at the level of the individual campus, public or private.
VI. Projections of Incomes Expenses, and Institutional Financial Viability
The disaggregation of expenses and incomes of a college better enables us to make guesses regarding financial trends, and the overall financial health of higher education. Let us turn first to the four principal components of income (discounting borrowing as only a short-term source of cash, rather than a fundamental source of revenue).
Tuition and fees. Tuition in the private sector, may be expected other things being equal, to increase at least as fast as prices in general if we merely assume that higher education should be no less subject to inflation than other goods and services. It may be expected to rise higher if we recognize the fact that higher education, like other services produced by the "productivity immune" sector, is one of those contributors to the average price rise which ought normally to be above average. Perhaps the rate of increase in family income--which ought to increase at the rate of inflation plus the average real growth in the economy, or precisely our predicted rate of increase in costs and prices for higher education--ought to be an appropriate and obtainable upper limit on the expected rate of increase in tuition and fees.
The limitation on tuition revenue in the private sector is not, fundamentally, that tuitions cannot continue to rise at rates corresponding to reasonable rates of increase in unit costs, although many institutions will come across price resistance at those rates. Rather, the fundamental limits to tuition growth in the private sector are demographic, and competitive. The demographic factor is the decline in the sheer number of middle and upper middle income high school graduates whose families can afford these very high costs. It is primarily a factor of birth rates, and is not really amenable to "solutions" by better management or more aggressive marketing.
But the problem of insufficient demand is exacerbated by the growing competitiveness of the nation's public institutions of higher education, which are offering much the same product at significantly lower tuition. The price disparity does not affect so adversely those private institutions, however pricey, that are "selling" mainly prestige, which is almost always costly. Nor will the disparity badly damage those private institutions that occupy special market niches, either programmatic (e.g., some health-related programs not available elsewhere) or social/cultural (e.g., a pervasive religious or ethnic atmosphere). But it is beginning to affect those that are not able to differentiate themselves from public competitors and that do not offer any greater prestige (and perhaps less) than their public counterparts.
These private colleges will have to moderate their tuition increases and also apply greater caution to the practice of price discounting in the form of institutionally-given financial aid. They must therefore compete on the cost side: reducing labor costs by substituting low-paid, part-time faculty for full-time faculty, and eliminating academic programs that fail to attract sufficient student demand and net tuition revenue.
In the public sector, tuition and fees tend to be set according to political or other criteria having little to do with the underlying cost of the product. Increasingly, public tuition rates are set last, generally to close the gap left by the insufficiency of state tax funding. Or, if the governor and the legislature believe the public university to be insufficiently "productive" (particularly when measured by volatile indices like average teaching loads), or otherwise resistant to changes occurring elsewhere in the state's pubic sector, they may deny the tuition increase as well as the tax support, and permit the resulting financial squeeze to force the public college or university to downsize, restructure, or alter teaching patterns in the ways sought by the political powers.
Government is the second principal source of revenue. Normally, revenue from government can be expected to increase as the government's own source of funds--tax revenues--increases, and as the public sector generally absorbs a greater and greater proportion of the country's productive capabilities. However, all is not well in the public sector, and higher education is no longer even relatively favored. There is a serious and growing sentiment against big government, and even more so against big taxes, And for what additional taxes revenues might become available, there is a long queue of public needs ranging from better elementary and secondary education, to job training, to environmental cleanup, to economic infrastructure, to more public safety--all of which await funding when the health of our public treasuries improves.
Arguably, higher education's problem is not that it is seen as any less important, either to the individual or to the larger society, but that the electorate is not convinced yet of the need for increased spending in this sector. This view is reinforced each time higher education "takes a hit," whether in the form of reduced federal aid, particularly damaging to the private sector, or reduced state support, in the public sector, and institutions seemingly continue to operate as effectively as before. Institutions of higher education, public and private alike, are enormously robust, at least to the public eye. There may indeed be losses in outputs occasioned by the diminution of governmental support. Colleges and universities may be loosing quality faculty, which will inevitably lessen their scholarly prestige some years in the future. They may be depleting their physical plant through deferred maintenance, which again will only become truly apparent some years in the future. Or they may be losing their one-time selectivity, which will lessen their prestige--but again, not immediately. In all these ways, colleges and universities can indeed be losing institutional vitality, and even viability, while seemingly balancing their books and maintaining most of their enrollments--and thus unintentionally encouraging governments to continue the squeeze until something more visibly dreadful happens.
A third source of income, principally for a handful of well-endowed private institutions, is the return on invested funds. This return, in normal times, ought to increase in proportion to both inflation and rising productivity--precisely in accord with our projected rate of increase in higher education's unit costs. The mid 90's are seeing good performance inmost equity portfolios. These can also go bad. But the main limitation to this source of income is simply its relative insignificance, except for the handful of colleges and universities that are already the wealthiest.
The fourth source of revenue that must at least keep up with the increases in unit costs is current giving. The largest percentage increase in private gifts and bequests has been occurring in the public sector, where the base was so very small, and any funds raised generally went for designated purposes or financial aid, and not to the general operations of the institution. Also doing well in fund-raising have been those with the largest and wealthiest alumni--not surprisingly, the very institutions that are probably doing best with the other two sources of revenue.
Annual fund raising for small and/or new institutions is difficult The competition for the philanthropic dollar is increasing . And we do not necessarily think to put $1.04 into the church collection plate just because we put $1.00 in last year and inflation since last year was 4 percent. In short, annual giving will continue to be lifeblood for most private institutions, and will increase in importance for public ones. But it is not likely, for most institutions, to offset losses from declining enrollments and state tax support.
Each of these revenue sources is limited for some or most institutions. The hugely endowed private universities may be the exception and may do well on all counts. Their endowment are generally well-invested. Their alumni base is affluent, and has done well in the last decade, during which the well-off generally became more so. Their applicant pools are deep, and most are able to afford high tuition. And after the parental contributions have been maximized and their endowed aid funds depleted, their student bodies are more willing than others to borrow.
Less-selective and less well-endowed colleges will have a much more difficult time meeting net tuition revenue goals. The combination of unfavorable demographics plus competitive disadvantage means that net revenue producing enrollments may decline. Declining enrollments will put all the more pressure on the financial aid and recruiting budgets, while discouraging the kinds of tuition increases that filled the gaps in the past. Revenues fall short of necessary expenditures. Reserves and other funds that are legally unencumbered will be spent and used for collateral for bank loans. At such time as there are no unobligated funds left to either liquidate or to collateralize another bank loan, the unluckiest of these institutions will be effectively bankrupt. And although colleges and universities are enormously robust, and although alumni, friends, and other saviors have intervened to save many a college that appeared to be on the financial brink, it is almost inevitable that there will be some actual closings in the year ahead.
Meanwhile, the public sector may benefit some from their edge in price competition, but they, too have suffered from the declining base of traditionally well-prepared high school graduates. Although these numbers are no longer plummeting, there is no boom in sight--at least not for students conventionally prepared for college.
The larger public research universities will continue to attract academically strong students, especially at the graduate and advanced professional levels. They will also continue to attract research dollars, although tighter federal research budgets, the rising costs of research, and the inevitable tightening of federal research overhead recoveries will lessen the beneficial financial impact of sponsored research to the university as a whole. The larger and more prestigious state universities will also continue to improve their private giving records. (By 1993, eight of the top 20 institutions in private gifts and grants were public universities.)
But some of the smaller state colleges with neither the location, prestige, programs, or capacity for sponsored research will be squeezed between declining public tax support and demographics, which in some states will be stagnant, and in others will present what is being called in California another "tidal wave" of college demand. Politics and public sectors being what they are, we are likely to hear such nostrums as "merge," or "restructure," or "consolidate," as governors, legislators, trustees administrators, and faculty cope with the new economics and finance of higher education.
(This paper was first written in 1977 for a graduate class in the economics and finance of higher education at the University of Pennsylvania; it was last revised in 1996.