Museum of unintended consequences/Grading oneself

From CSWiki

< Museum of unintended consequences
Revision as of 06:19, 28 August 2009 by Russ Abbott (Talk | contribs)
(diff) ←Older revision | Current revision (diff) | Newer revision→ (diff)
Jump to: navigation, search

The United States military buys everything from bullets to satellite systems from the private sector. Production of these devices and systems is sometimes quite complex—especially in the case of large and complex items such as satellite systems. Managing the development of a large and complex multi-billion dollar system often takes year, sometimes a decade or more. Typically a military officer is in charge, with one or more contractors doing the actual work. These efforts are called programs. Most people have probably heard of the GPS program in which satellites have been launched that are capable of providing ground GPS receivers with precise enough information that the receivers can figure out where on earth it is.

In the course of the development of a program, the military overseer is required to grade the contractor on how well he is doing. Unfortunately, the military officer is himself (or herself) graded on the basis of how well the program turns out. If the military officer can show that the contractor is doing a good job, that reflects well on him or her as the manager. So the military program manager is tempted to cover up contractor problems and simply give the contractor a good grade.

This temptation is exacerbated by the fact that most military assignments are typically limited to a few years at most. If a military program manager expects to leave a program before problems that he or she observes cause any actual damage, he or she is sorely tempted to do his or her best to cover up the problem. Furthermore, the program manager can be fairly certain that the problem won't be visible outside the program until sometime after he is she has left, by which time it can be blamed on someone else.

When the problem becomes visible, the person in charge at the time can blame his or her predecessor. But that blame will not actually be recorded on the predecessor's record. So no one is actually hurt.

Thus the rules that determine how program managers are evaluated and how long they remain with a program have the unintended consequence of encouraging the manager to cover up problems in the program.