Attrition: Bad Numbers Hide the Unexpected

Archives

March 27, 2007: U.S. Army bureaucrats screwed up some statistics again. This time it was the desertion rate. No big deal. The army had 3,196 troops desert in 2006, 2,543 in 2005, and 2,357 in 2004. The desertion rate hit a peak in 2002, with 4,483 walking away. The army began to screen more carefully for adaptation problems, and has cut their number of deserters nearly in half. Desertion is the largest cause of losses in the military, larger than combat, and non-combat, deaths and serious (resulting in medical discharge) injuries.)

The counting error, attributed to mistakes made by several clerks and supervisors, did not change the total number of deserters for the period2000-2006. Thecorrected number is 22,468, compared to 22,586 for the bad count. Most of the errors were deserters being listed in the wrong year.

The draft ended in 1972, and since then, deserters have largely resulted from volunteers who had problems adapting to military life. A deserter is anyone on active duty that is away from their unit, without permission, for more than 30 days. The military doesn't go looking for deserters, but instead alerts police throughout the nation. If a deserter has any encounters with the cops, the desertion will show up, and the deserter will be arrested and turned over to military police. The deserter is then returned to their unit, where the punishment ranges from loss of rank and dishonorable discharge, to that, plus up to five years in prison. The most common punishments are at the low end, although in the last few years, there have been more cases of deserters being given another chance to complete their enlistment.

Those arrest warrants for deserters never expire, and some Vietnam era deserters are still getting picked up. They get the same treatment as deserters of more recent vintage. The military never expects to completely eliminate desertion. Despite increased efforts to keep potential deserters (usually the less educated and from broken homes) out of uniform, the rate is expected to go up again once the fighting in Iraq and Afghanistan is over. The current desertion rate, about .5 percent of the force, is much lower than the peak year for Vietnam era desertions, when 3.4 percent of the force took off.

Bad numbers are nothing new. For example, the Department of Defense used to report that the number of combat deaths in the Korean war were higher (by over 10,000 dead) than they actually were. This was because, early on, someone mistakenly added all the accidental deaths, world-wide, for the United States military, during the period of the war (1950-53), to the total combat dead. It wasn't until the 1980s that this got cleared up.

And then there are some bad numbers that will never be cleaned up. Friendly fire incidents in past wars were routinely misreported, usually at the lowest levels (friends of those who got shot, or did the shooting.) Any attempts to get to the bottom of friendly fire statistics from old wars, would open too may psychological wounds. Same with the misreporting of dead soldiers as "missing in action" during World War II. This was often done by the dead soldiers family, so the widow could collect the soldiers pay (which was higher than widows benefits) for a while longer. There are a lot of bad numbers out there, and an interesting story behind many of them.