Activity analysis and cost analysis in medical schools
Journal of Medical Education
There is no unique answer to the question of what an ongoing program costs in medical schools. The estimates of program costs generated by classical methods of cost accounting are unsatisfactory because such accounting cannot deal with the joint production or joint cost problem. Activity analysis models aim at calculating the impact of alternative combinations of school activities. As now practiced, activity analysis yields the incremental cost of an activity given all other school activities,
... school activities, that is, pure program costs. Such analysis is a potentially useful management tool. Input/output analysis is a special form of activity analysis in which some activities provide both final outputs and input into other activities. Following the introduction in Section I, this paper deals with cost analysis in section II and III. Section II discusses cost allocation studies that use the tools of classical cost accounting. Section III analyzes the use of linear regression as ABSTRACT In circumstances of joint production there is no unique answer to the question of what an ongoing program costs. One can only speak of "pure" and "joint" program costs, the sum of "pure" costs being less than total cost and the sum of "pure" and "joint" costs being more than total cost. This need not be a source of frustration to the policymaker, however. If the policy question is whether a new .program is "worth doing" the appropriate cost concept is the sum of "pure" and "joint" costs. If the policy question is whether an ongoing program is "paying for itself" the appropriate cost concept is "pure" cost. It must also be asked whether the institution is "paying for itself," however, for the answer to this question is not implied in the answers to the questions of whether individual programs are "paying for themselves." A major modification of classical cost accounting procedures is required if appropriate cost estimates are to be obtained. Activity analysis models could provide the basis for such estimates if the models were constructed so as to capture the phenomenon of joint production. The input-output model does not, and hence is unsatisfactory.