Admiral Rickover on improving weapons acquisition

Today's quote is from Admiral Hyman G. Rickover's memo to David Packard on May 22, 1970, concerning Packard's 5000 series directives to improve the acquisition process:

“... to bring about real improvement in weapons acquisition cannot be corrected by management policy directives… My experience has been that when a directive such as the one you propose is issued, most of the effort goes into the creation of additional management systems and reports and the preparation of large numbers of documents within the Service to “prove” that the requirements of the directive are being me—in order to justify funds for the Service…. I think that if you check you will find that virtually all of these documents, which have required so much effort to prepare, are used almost entirely to gain approval for programs, and are simply filed away once funds are released. They are of no use to the working people in performing their jobs. In my opinion the present situation in this regard is worse than I have ever seen it. Further, I predict that the way the present bureaucracy would implement your proposed directive will make matters still worse.”

Rickover was prescient on this matter; inevitably any streamlining and decentralizing initiative gets drowned out by the bureaucratic structure. Yet perhaps that is for good reason, because true deregulation simply cannot happen in today's industrial culture due to the wastefulness and deception contractors can get away with.

In some ways, the statement "trust, but verify" makes a lot of sense in defense acquisition. There must be a building of trust between a multiplicity of actors, not just a monolithic government and a handful of firms. But for trust to solidify, there needs to be verification of that trust -- namely understanding the value of project results.

If the purchaser can properly evaluate a project outcome, which only comes after years of experience with numerous similar projects, then trust is a matter of after-the-fact evaluation followed by reward or punishment (generally filtered through reputation effects, not legal recourse).

If the purchaser knows relatively little about evaluation, evident by constant information gathering and justification requirements, then trust may appear to be a matter of ensuring proper accounting and regulation of profit levels.

The problem with this story, however, is that no one was better than Rickover at understanding technical aspects of engineering projects, and Rickover still relied on cost accounting standards to reveal fraud or abuse.

Though fraud and abuse certainly did occur, it was because of Rickover's other complaint, that the lower echelons were losing technical talent fast and there wasn't a cultural knowledge of project evaluation, and more importantly, they couldn't make project decisions without getting funding requests through 30 or more offices at all levels of the Department. 

Nothing more than this can stamp out the diversity of organizations and projects leading to what is most important: the increased information processing capacity of the Defense system devoted to "doers" and not "reviewers," as Rickover would say.