The Power and the Money

“Frankly, I’d like to see the government get out of war altogether and leave the whole feud to private industry.”

Thus spoke Major Milo Minderbinder in 1943.

Was he right? Does contracting out combat support functions save money? I mean, that is not quite the whole feud, but it is some of it.

The best answer comes from study published by John McGrath, at Fort Leavenworth. Below are the figures from Table B-2 for various American wars. The figures apply only to the Army but rightfully include staging areas in Korea and both Gulf Wars: Japan and Kuwait, respectively.

So what do we got? At first glance, success! A higher tooth-to-tail ratio, awesome sauce!

We are back to the levels of World War 2! (AEF = American Expeditionary Force of WW1; ETO = European Theater of Operations in WW2.) Hooah!

Only, well, that does not include the contractors. We really should put them back into the mix, since they did jobs that soldiers did in Gulf 1 and before.

Ah. Well. Not so good.

Now, to be fair, these figures do not include the extensive host-country support the U.S. received in Korea, which would make that war look pretty bad. (Korea would fall to 28% combat troops on a BOTE calculation.)

They also do not include potential cost savings. After all, if private contractors are cheaper, then you would expect that the military would use more of them. Calculating that, however, is incredibly complex, because American operations have become far more capital and resource-intense over time. The simple calculations show ever-rising real deployment costs, but that could be a reflection of the fact that the American soldier in Gulf 2 was rather more lethal (and well protected) than the American soldier of Gulf 1, who was both more lethal and better protected (and better fed) than the American soldier in Vietnam, who was more lethal than the American soldiers who beat Hitler.

It is hard to figure.

But the headline figures should induce skepticism.

This, note, is just for outsourcing support functions; we have not gotten into questions of outsourcing actual genuine combat. You know, what Major Minderbinder really wanted to see. That is an even less-satisfying topic for another post.

Selection bias

 There are a lot of papers in the “history only happened once” genre, where somebody examines an event that happened hundreds of years ago and proclaims that the effects persist today. Some are truly excellent and tell us things we did not know. They find a correlation most did not expect, use unassailable data, and have a testable explanation of why the effect persists. Some are very neat but may have other explanations. Some are trivial: they use fancy methods to reaffirm persistence that had an obvious origin and is well-known to even casual observers. And some are rather contentious: the data and results are not as clean as the authors would claim. 

Now we have a paper claiming that it matters whether an area fell under Dai Viet rule in Vietnam before 1833. The Dai Viet state, the paper claims, was organized along Chinese lines with strong village councils. Those areas today (and under South Vietnam) were places where “citizens have been better able to organize for public goods and redistribution through civil society and local government.” 

I have to wonder: what happens to all the papers that find no effect from long-ago events? Other than the ones on strategic bombing, they never make it to publication. What would a meta-analysis find? Maybe there really is no there there.