Some confusion appears to exist about the relationship between quantitative analysis and subjective judgment in the practice of systems analysis. Partly responsible is the improper identification of operations research as systems analysis. In operations research, the objectives are given, assumptions about the environment are specified, and so forth. The task is to calculate an optimum solution for a fixed level of resources. Enthoven points out that although the method is similar, the perspective of systems analysis is broader:
“[Systems analysis]… analyzes alternative objectives and explores their implications. It is focused more on exploring the implications of alternative assumptions than on analyzing in extensive detail the implications of a single set of assumptions. Systems analysis ordinarily is not concerned with computing an optimum solution.”
Systems analysis does not utilize sophisticated mathematical techniques, as does operations research, but rather emphasizes basic economic concepts such as marginal products and marginal costs. Enthoven once observed that one learns the necessary economic theory in sophomore classes but not until after one’s Ph.D. is acquired does one believe what one has learned and acquire a vested interest in marginal analysis.
That was Clark Murdcock’s excellent book, Defense Policy Formulation (pp. 56-59). Required reading for those interested in acquisition.
I love that line. Enthoven, who became director of the Office of Systems Analysis at only 31 years old, joined RAND right out of his economics classes at MIT. He actually thought that you could simply apply sophomore econ to solve defense problems. I suppose the naivety isn’t necessarily apparent until it is actually tried out.
I remember in sophomore undergrad micro-econ, studying marginal analyses. I wondered how this was actually used in firms to decide upon the optimal course of action. I assumed that it was used in management, I just didn’t know how. It took a few more years for me to come across ideas like Armen Alchian’s Evolution, Uncertainty, and Economic Theory (1950), which completely dismantled the idea of “profit maximization.” I think the same goes for planning the “optimal” weapon system. I find it useful reading that paper every couple years.
It’s unfortunate that Enthoven and others at RAND simply didn’t look at the accounting literature to realize, when firms tried marginal pricing in the first part of the 20th century, they badly mis-priced goods. It did not lead to their survival in the economic ecosystem. Firms ended up pricing at average cost to stay alive.
This discrepancy between marginal cost theory and average cost practice led to major economics debates, and eventually, got Alchian to write his paper. He was flabbergasted that economists had so thoroughly forgotten competition, evolution, and the purpose of profit and loss.
The Enthoven quote comes from “Defense and Disarmament: Economic Analysis in the Department of Defense.” In the American Economic Review 53 (may 1963): 287-301. It’s kind of amazing to think about defense work getting into top journals like the AER. That would be highly unlikely today. But times have changed.
Here’s some more from Murdock:
Systems analysis, therefore, was intended to take advantage of the new information categories the programming process created. Activities were to be related to functions, and systems analysis was to aid in the allocation of resources within programs, among programs, and between defense and other needs. These were the two primary management innovations.
The philosophy of incrementalism was itself a reaction to the comprehensive rationality of the rational-deductive model of decision-making utilized in economic theory. This model posits a rational decision-maker with a clearly defined hierarchy of goals. In any decision-making situation, the decision-maker knows all the alternatives and their consequences and chooses among the alternatives according to criteria based on his ordered goal system. The incrementalists, however, found this model unrealistic – not only descriptively but also prescriptively – in view of the analytic complexities of decision-making and the limitations of men as problem-solvers. Of necessity, man’s problem-solving capabilities are limited by the amount of information he can collect and evaluate. His goal system, moreover, is not ordered but rather a myriad of conflicting, often unarticulated, goals, which are attended to sequentially. The search for alternative and the calculation of their consequences is a costly, time-consuming process that prohibits any near approximation of complete knowledge.”
Leave a Reply