Part of the reason for “programming” the budget was to give decision-makers at the top the ability not only to decide on how much money the departments would spend, but exactly on what projects and objectives they would accomplish. With a new method of civilian control over military decisions, Alain Enthoven and other “whiz kids” under Robert McNamara thought the information would be used to arrive at better decisions than the military could have muddled through using rules of thumb.
It turned out that the information, which required so much coordination, was of not much use to the top civilians. As Enthoven complained in his 1971 book, How Much Is Enough?:
If the Congress shouldn’t have to make such multi-billion dollar decisions, who should? Senator Mundt’s attitude is representative of a serious problem faced by the civilian leaders of the Defense Department during the years 1961-1968. PPBS was making available more useful information to Congress, but ironically some members often didn’t seem to want it.
They didn’t want it because what use is the program budget information if they don’t understand the context behind all the analysis that went into it.
The belief at the time was that top policy makers and administrators did not need to have long experience in military or technical matters to make good decisions. All they needed to know were the relevant variables and have available the requisite data to find the single-best solution to any problem. The answer was objective and could be verified by third-parties.
Yet as we have come to learn, control of complex operations cannot be managed “scientifically” by transients. Most presidential appointees, let alone Congressmen, cannot have the relevant knowledge to make good use of such highly aggregated information that the PPBS produces, especially since they are predictions of the future.
Technical choices are often best made by the person with the most knowledge, usually an experienced “line” manager or officer. Accountability for those choices is important after-the-fact, such as through operational tests of equipment.
The decision process in a complex system is better described by selecting the right people first, who then grope towards the correct program decisions, rather than today’s process of deciding on program specifications, then assigning some interchangable person to go manage it.
Leave a Reply