David R. Henderson  

Sunk Cost in War

PRINT
Quietism and the Bubble... A Non-Keynesian Turing Test...

I am used to people making the sunk cost fallacy when discussing war, that is, in one of the most important cases in which not to commit the fallacy. So I was pleasantly surprised by a segment on The O'Reilly Factor last night. One of O'Reilly's regulars, when he wants someone to discuss foreign policy, is a hawkish retired Colonel named Ralph Peters. Here's how hawkish he is: he advocated that the U.S. government murder Julian Assange.

O'Reilly, pointing out that things seem to be breaking down in Afghanistan, asked Peters what he would do if he were in charge. O'Reilly set it up by mentioning all the "blood and treasure" that has been lost so far in Afghanistan. Peters answered that the U.S. government should pull out the U.S. troops. (Peters also wanted to keep some U.S. soldiers and drones there.) O'Reilly said words to the effect, "But then all that blood and treasure would be wasted." Peters answered, "But you can't get it back by spending more blood and treasure." I've seen Peters a number of times and have never been impressed with his intellect. But this time I was impressed with his clear thinking.


Comments and Sharing


CATEGORIES: Cost-benefit Analysis



COMMENTS (8 to date)
Xerographica writes:

What a coincidence! I was fortunate enough to watch this bit of the show and wrote down that same exact quote from Ralph Peters!

Not sure if I got it exactly right but I believe he said..."When we invest our blood and treasure I expect a positive return"

It's been a few years since I was stationed in Afghanistan, but looking through my photos in the folder which I dedicated to "crops"...I see figs, grapes, pomegranates, wheat, eggplant, etc...but by far the most common crop was the opium poppy. If a small village only had one crop then it would invariably be the opium poppy.

Everybody wants the most bang for their buck. It's a universal truth. Yet...I struggle to convey to people the value of allowing taxpayers to directly allocate their taxes.

If we were able to directly allocate our taxes then, given my partial knowledge, I would certainly allocate a portion of my taxes towards our efforts over there. It would also make it worth the effort for me to share my partial knowledge with others.

liberty writes:

Keeping in mind that I am not expressing my own view s here at all, I think the 'sunk cost fallacy' accusation, about most, is a bit unfair. Most of those people who argue that we've already lost blood and spent treasure there and should stay are not saying that we should stay simply because we've already sunk those costs in and therefore we should stay indefinitely, or until it's complete like the building of a building, with some known end, just because we started.

They are saying instead that IF we pull out, the lives and money will have been wasted, but IF we stay, we might be able to win, and then the lives and money will NOT have been wasted. They are saying, essentially, that the project is half-finished and when half-finished it is useless but when fully-finished it is NOT useless. This is not a sunk-costs fallacy, it is a debate over whether they are right or wrong about the possibility of success.

Ken B writes:

Of course if we are close to success, and only a little more is needed, then it's not the sunk cost fallacy. So let's grant ad arguendo that we are no closer to success (however defined) than X years ago (any X > 0). O'Reilly's remark resonates with people because of notions of honor and keeping faith. Cutting our losses is *felt* as a betrayal of a sacrifice. This is why it's important to have a symbolic victory of some sort before you 'declare victory and go home.'

David S writes:

Ken B said:

"Cutting our losses is *felt* as a betrayal of a sacrifice."

Whenever someone says "felt" about morals, it is worthy of investigation in my opinion. Humans have had many thousands of years to finely hone "feelings" about morality, and most are based on a need that can be discovered.

In this case, for example, I believe that strategic game theory would lead to staying in the war longer than a purely local cost/benefit analysis would suggest - even if you are losing the war. There is a hidden objective in war, beyond winning the current war - preventing the next war. If you immediately give up as soon as the war's net present value goes negative, your future adversaries will fight harder than if they know that you will never give up.

So although you have to apply a steep discount rate for the effect on future wars, I believe that the effect still would significantly move the point at which you should give up in a struggle. Honor and duty have their advantages.

Ken B writes:

@David S:
I don't disagree with anything you said.

I also think DRH and others seriously understimate the cost of withdrawal. We are witnessing a civil war within Islam. Our role is secondary, mostly providing a target for the back-to-basics bunch to score points off of by humiliating, or to provide secondary support for the modernizers. It is very costly to allow the b-to-b crowd to score high profile victories because that will likely have a big effect on the internal struggle.

Of course staying and getting beaten is bad too, and can be worse if we double down and then quit, so this does not by itself show we should stay. But it's an extremely important point usually ignored by the let's-pull-out side.

David R. Henderson writes:

@liberty,
They are saying instead that IF we pull out, the lives and money will have been wasted, but IF we stay, we might be able to win, and then the lives and money will NOT have been wasted. They are saying, essentially, that the project is half-finished and when half-finished it is useless but when fully-finished it is NOT useless. This is not a sunk-costs fallacy, it is a debate over whether they are right or wrong about the possibility of success.
If that’s what they’re saying, then, yes, you’re right that it’s not a sunk-cost fallacy. I think O’Reilly wasn’t saying what you generously took him to say. But I could be wrong. What was clear to me, though, was that Ralph Peters, who I would have expected would make the sunk cost fallacy, so clearly did not.

Kevin writes:

David I share @liberty's possibly generous understanding of this expression. I think people who say this sort of thing have intuition that tells them that there is a "value" of the current situation that they estimate at the cost to recreate it, which they further intuit to be equal to the cost incurred thus far. They also intuit that this is the value that will be lost if the project is abandoned. Thus the expression.

It's too bad more people don't intuit, among other things, the continuing liability in blood and treasure of preserving and building on the status quo.

ilsm writes:

“If we would just take the profit out of war there wouldn’t be any.” Woody Guthrie

A discussion of the sunk cost fallacy, found while I was working in a failing DoD weapons program. http://www4.ncsu.edu/~sgallen/Of%20mice%20and%20economics.pdf

The sunk cost fallacy has been used to keep building failed weapon systems for about as long as there has been money to be made in selling shoddy things to the government. Unfortunately in the US DoD it is usually used by the government guys who want jobs a few years down the road, working on the franchises for the failed weapons they managed as government workers.

The paradigm of the sunk cost seems to be to decide to waste more, new resources so the guys making the mistakes, and hiding them are not embarrassed.

For companies making money on Afghanistan or a failed weapon system the sunk cost fallacy is marketing.

Comments for this entry have been closed
Return to top