For some problems, great gains will come from unique binding choices resulting from systems analyses; for others the gain will come from diversity of actions… At the risk of some violence we can distinguish two kinds of choices: (1) a choice of a single optimal action; (2) a choice of a diversity of actions such that subsequent selection may be made in light of revealed developments. The first involves a longer range fixity, while the second involves a shorter range bind but purchased at the expense of higher current costs. In what situation is the latter principle of diversity preferable? And in what situations is the former appropriate? Do systems analyses help us to answer these questions? Does it help us select the diverse or unique actions? We think the answer is yes to both questions, but the point we would like to see emphasized in the course is that systems analyses also have the potentiality of misleading us.
That was from Armen Alchian and Reuben Kessel’s 1954 classic, “A Proper Role of Systems Analysis.” This is at the heart of current debates on acquisition reform. The predominate view is that a single optimal action should be chosen such that, before the thing has been prototyped, officials have a good idea of all the risks, annual cost estimates, criteria for testing, depot activation issues, and so forth.
However, the principle of diversity of action is more expensive in the near-term because it creates options down the line. Not everything can go into production, but it provides insurance against changing states of the world, whether technologically, in the threat, or otherwise. Indeed, the more unpredictable the world gets, the more likely a project gets “selected” by the environment as if by chance rather than through rigorous planning. Here are the authors:
The insurance principle of diversified investments in development is superior to the principle of developing and procuring one flexible weapon. This assertion is refutable. But so strong is our conviction in this, that we strongly recommend this theorem as a basic part of the systems analysis. In all frankness, we are obstinately insistent that this is true for research and development decisions; we are of an open mind on the issue of whether or not it is true for procurement and other categories of decisions.
As weapons become more and more defined by software, breaking down the barrier between well-structured procurement/operations decisions and more open-ended RDT&E decisions, the greater will be the benefit from the diversification principle.
Leave a Reply