Economic Evaluation in Education. Henry M. Levin
assessing costs is the correct use of the terminology of economic evaluation. When terms are used loosely, it is very difficult to explain and interpret costs. Thus, this chapter focuses on defining and clarifying terms. This discussion will provide a foundation upon which to design and apply the ingredients method as described throughout the remaining chapters of the book.
3.1. The Concept of Costs
Costs refer to the value of all the resources used to implement the program in question. Every intervention uses resources that could be utilized for other valued alternatives. For example, a program for raising student achievement will require personnel, facilities, and materials that can be applied to other educational and noneducational endeavors. If these resources are used in one way, they cannot be used in some other way that may also provide useful outcomes. The human time and energy as well as the buildings, materials, and other resources used in one endeavor have other valuable uses. By devoting them to a particular activity, we are sacrificing the gains that could be obtained from using them for some other purpose.
If we assume that a specific intervention will have an effect that is found in a good effectiveness study, we want to know what it costs to replicate that intervention to compare its effectiveness. The cost is represented by all of the resources or ingredients required to replicate the program, no matter how they were financed or provided. Virtually all resources have a cost, even if they were provided in kind.
Accordingly, the “cost” of pursuing the intervention is what we must give up by not using these resources in some other way. Technically, then, the cost of a specific intervention will be defined as the value of all the resources that it utilizes had they been assigned to their most valuable alternative use. In this sense, all costs represent the sacrifice of an opportunity that has been forgone. It is this notion of opportunity cost that lies at the base of cost analysis in evaluation. By using resources in one way, we are giving up the ability to use them in another way, so a cost has been incurred.
Although this may appear to be a peculiar way to view costs, it is probably more familiar to each of us than it appears at first glance. It is usually true that when we refer to costs, we refer to the expenditure that we must make to purchase a particular good or service, as reflected in the statement, “The cost of college was $5,000.” Here, the statement refers to expenditure on fees and tuition the college charges students who enroll. In cases in which the only cost is the expenditure of funds that could have been used for other goods and services, the sacrifice or cost can be stated in terms of expenditure. However, in daily usage, we also make statements such as these: “It would cost me a full year without working to attend college,” or “The time cost of being in class is too high.”
In each of these cases, a loss was incurred, which is viewed as the value of opportunities that were sacrificed. Thus, the cost of a particular activity is viewed as its “opportunity cost” and should include all lost options. Of course, this does not mean that we can always easily place a dollar value on that cost. In the case of losing a year of work, one can probably say that the sacrifice or opportunity cost was equal to what could have been earned; a similar logic applies to missing class, even as the actual money value of time for learning might be harder to estimate. These costs should be added to the extra money the student has to pay in fees.
The most common method for placing monetary values on resources is that of using their market prices. According to economic theory, when a market for a particular good or service is perfectly competitive, the equilibrium price established by that market will represent the value of that good (Mankiw, 2011, Chapter 7). The market price reflects the point at which the demands of purchasers and the supplies of providers lead to transactions that clear the market. This is known as market value. These participants are basing their willingness to buy or sell on their own opportunity costs, so the market price should directly reflect opportunity cost.
However, market prices are not a very accurate reflection of opportunity cost in distorted markets. There are many reasons why markets might be distorted. These include highly concentrated markets with less competition, information failures, and externalities. In education systems, monopoly features are common. For example, the cost of hiring a teacher is often based on district salary scales or statewide certification requirements on the demand side and based on collective bargaining contracts on the supply side. These supply and demand distortions may move teacher pay away from what might be considered as a competitive market price for teaching services (see Eberts, 2007; Jacob, 2007). However, teachers can select alternative occupations, and districts/schools must pay competitive wages or their teachers will leave for other districts or schools or for other employment. Generally, the distortion in teacher labor markets might be small such that the market wage is appropriate. Nevertheless, for each input, it is important to consider if there are distortions away from market prices, how large these distortions might be, and if they apply to the decisionmakers’ context.
For some resources, market prices may not be available. When attempts are made to ascertain the value of a resource that does not have a competitive market price, the estimated value is called a shadow price (see Boardman, Greenberg, Vining, & Weimer, 2011, Chapters 16–17). Shadow prices should be based directly on opportunity cost. For example, a school district may decide to lend an old facility to a new program. There is no financial transaction, because the building was purchased and paid for a long time ago. Moreover, no market exists for this type of facility; there is no direct market price, so we must impute a shadow price valuation although the facility may have some lease value to nonschool users. This resource must therefore be valued based on what is would cost to maintain the facility in its current state, or to demolish the facility, or to refurbish it for alternative use. Another more common example is volunteer time: By definition, volunteers are unpaid, so there is no direct market price. Volunteer time must therefore be based on the opportunity cost of the volunteers’ time.
Both market and shadow pricing approaches may be used for purposes of cost estimation. Both approaches are derived from the idea of opportunity cost, although we recognize that opportunity cost is a subtle concept that requires careful interpretation. Imagine a school with a gym that is typically unused in the summer. A nonprofit agency proposes a summer school basketball program for high school dropouts. Claiming that the gym is empty, the school might think that the opportunity cost is zero. However, a zero opportunity cost is rarely appropriate because the facility has other uses that represent an opportunity cost if not exploited. It is unlikely that there really is no other alternative use of the gym (e.g., for a different program or leasing for nonschool activities). Also, using the gym in the summer will likely necessitate maintenance costs later. Finally, the assumption of zero opportunity cost might be invoked in an ad hoc way for any resource (e.g., for example, the assumption that the cost of parental time contributions are zero because the parents would otherwise be relaxing at home). Potentially, the analyst could claim that a lot of resources had no alternative use and hence assert that the costs of the program are almost nothing. The conventional—and correct—assumption is that all resources have a positive opportunity cost, as all are needed to replicate the program regardless of financing.
3.2. Cost Per Unit
It is important to be clear about what we mean by costs of education (for a parallel exercise for costing out prevention programs, see Foster, Porter, Ayers, Kaplan, & Sandler, 2007). Strictly speaking, in economics the term cost refers to “cost per unit of output,” for example, cost per car manufactured, cost per 1,000 computers produced, or cost per medical consultation. With this definition, an enterprise that has a lower cost per unit of output is more cost-effective (efficient) than one with a higher cost. This definition corresponds directly to the terminology of a standard economics textbook: Total cost is the cost to produce a given quantity of output, average cost is the cost per individual unit of output, and marginal cost is the cost per additional unit of output. These definitions work well for production decisions.
For educational services, the term cost is slightly different. It refers to the value of all the resources needed to deliver the intervention to the intended population. The cost term is therefore cost per participant, per cohort,