August 05, 2002
The Profits Of War

I'm no economist, but it seems to me that Will Wilkinson of The Fly Bottle and Megan McArdle of Live . . . from the WTC are missing the point in arguing that war is definitely (Wilkinson) or possibly (McArdle) bad for the economy. Many economically useful activities aim not to make money but to avoid losing it. If I buy a piece of furniture for $100 and later, when I have no further use for it, sell it for $30 at a yard sale, I'm not being stupid: that's $30 more than I would have if I burned it or buried it in the backyard or gave it to Goodwill. (We'll leave out the possible tax advantages of the last as unnecessarily complicating my argument.)

To take a more pertinent example, is not the defense budget a form of insurance? I've spent more than $20,000 on car insurance alone over the last 30 years, without ever filing a claim. Am I too stupid to know that I've been throwing my money away? (I sure could use that money now.) Or was it worth it to avoid the catastrophic effects that would have followed if I had wrecked my car or someone else's and put myself or others in the hospital, as I might very easily have done? I think the latter is correct. Although I am extremely unlikely to put someone in a wheelchair or a coma for the rest of his life, such an action would be so extraordinarily expensive as to justify paying quite a lot for insurance to avoid a multi-million-dollar liability. Even a very expensive army that never fights a war can be a wise investment if it prevents war. If war turns out to be necessary, and expensive, the expenditure can still be justified if the alternatives are worse. We already know some of the things Al Qaeda and their buddies would like to do to us, and they certainly seem worse than war on Iraq.

I don't know how much the destruction of the World Trade Center, part of the Pentagon, and four airplanes has cost and will cost the U.S. economy, but I'm quite sure that it would have been worth spending another ten or twenty billion dollars on defense (broadly defined) over the last few years, if that would have sufficed to prevent that destruction.

So far I have only considered the economic costs of terrorism, not the deaths of so many people (except in so far as those are economic phenomena, loss of skilled employees and so on, where they can be counted in the economic cost). Again, I'm no economist, but I don't think economics can tell us how much to spend to avoid the deaths of ourselves or our loved ones or our fellow citizens. Most of us would have been willing to spend quite a lot to avoid those particular deaths, and are willing to spend quite a lot to prevent similar (and possibly worse) things from happening in the future. It's hard to think of any part of the federal budget that is more justified than defense, profit or no profit.

Posted by Dr. Weevil at August 05, 2002 01:02 AM

I doubt if anyone disagrees about the necessity of the spending. But I think most economists would tell you that money spent on any form of insurance is basically part of the cost of doing business--"the business" in this case being an existing and safe country. Without question it's an economic "waste" in the strict sense--if we could do completely without it we would, but of course in the real world that would be completely idiotic.

If you think of it in terms of physics, it is more easy to understand. In physics, you have a large amount of potential energy from any fuel source. However, in converting it to useful energy, you lose most of the potential, usually in the form of heat. We should think of spending on things like defense, or insurance, or repairs and upkeep on cars or homes or whatever, as "waste heat." Ideally, we get no waste heat because we're so efficient, but of course the ideal is nearly impossible. (If someone could invent the economic equivalent of fusion though... oh never mind I'm rambling).

Thus I think the argument should be obvious: defense spending is waste heat. It's also unavoidable because sooner or later the whole thing will break down without it.

Does it help the economy? If you are a Keynesian you say yes, because you believe any spending by government is helpful in some fashion. If you're an Austrian or Chicago school thinker, you'll say that it can have short-term positive effects but will inexhorably exert long-term negative effects---much like ramming the accelerator pedal to the floor so you can get a quick jump start, which works, but burns excess fuel and increases wear and tear on the engine.

How much better off economically would we be if we we spent not one penny on the military? Why, we'd do fabulously better. For about five minutes. It's otherwise a fantasy.

Hrmph. I think I should post this on my blog...

Posted by: Dean Esmay on August 5, 2002 10:43 AM

I wasn't missing the point -- that's what I was saying when I pointed out that military spending is indeed a higher valued use if the Canadians are massing at the border -- but trying to address a different point, which is that government spending on the military is somehow "good" for the economy independantly of its function of protecting that economy from destruction.

What's funny is that the Times is arguing that welfare spending has this magical stimulative effect, while war does not, even though both involve taking resources out of the productive economy and destroying them (welfare takes labor out of the economy, and a lost hour of labor can never be replaced; war, men and materiel). Sully is arguing the opposite, that war helps and welfare hurts. Both are equally internally inconsistent without the qualifier that there are limited circumstances in which government spending may help, but that those circumstances are so rare that it is hard to develop empirical standards for determining when one is in one, and what kind of spend remedies it.

Posted by: Jane Galt on August 5, 2002 11:05 AM

Maybe that's not the right question to ask. I think the question is: Can we afford not to topple Saddam Hussein and the rest of the rogue states and states aiding and abetting terrorism(Tehran, Damascus, Riyahd)?
What would be the effect on the economy if no war were instituted with Iraq and terrorism were allowed to proliferate? (I'll ignore the challenge that there's no proof that Saddam participated in 9/11--the proof of his involvement with Islamic terrorist groups throughout the nineties is too well established and beyond cavil and itself constitutes sufficient reason to grant him martydom and 70 virgins even if he had nothing to do with the World Trade Center.) A few more terrorist incidents (actually, one more would probably be enough), say, some biological agents in New York City's water supply or a dirty bomb exploding in the Port of Miami--how would that affect the economy? I'd damn well bet the market would plunge more precipitously than it has from Enron, Worldcom, et al. No more single-digit unemployment, etc., etc.

And anyhow, isn't this type of question somewhat perverse--a case of Hamlet-like neurotic paralysis? Forget the disutility and marginal- propensity-to-consume curves. I'll bet after being notified about the 2,000 some deaths incurred at Pearl Harbor, FDR didn't turn to Harold Ickes or one of the other Brain Trusters and clever young things who advised him and ask what the effect war would have on GDP; he simply and summarily declared war.

A society that's confident in its concepts of right and wrong and of in its right to survive and defend itself and visit destruction on those who would destroy it first sends its soldiers to strike first; a society that's lost its nerve calls in the accountants.

Posted by: Al C. Biadeez on August 5, 2002 01:30 PM

I respond at the Fly Bottle. I say military spending is not much like insurance when you think it through. But the military's most like insurance in the moral hazard problem it creates. Protection against risk increases the very risky behavior that creates the need for protection!

Posted by: Will Wilkinson on August 5, 2002 04:47 PM

I followed this line of comments from Jane's blog and I wasn't going to add anything but the logic of that last comment is so lame I had to put my 2 cents in.
Protection againist risk does not increase the risky behavior.
In particular with the military bigger is always better. Sometimes it is hard to see this because if, like the US, you have overwhelming power you tend to not have to use it.
In most areas you build things to use them the military builds things so they won't have to use them. The problem comes when your enemies(yes virginia we do have enemies) perceive that you will not put that power to use. This is as bad as not having the power at all.
A threat is only good if the threatened person thinks it is not a bluff.

Posted by: Starhawk on August 5, 2002 08:09 PM

Having a strong military, historically, is usually a strong deterrant to invasion. People don't fuck with you if they're afraid to. So it is a form of insurance.

The second problem is this notion that a military increases your odds of military adventurism. The very fact of having a strong military does not, ipso facto, make you more inclined to meddle in the affairs of other nations. It may make you less fearful of doing so, but by itself it does no such thing. Furthermore, just like insurance, if you engage in risky behavior your rates often go up; you get charged more for smoking with a lot of policies, if you get speeding tickets or drive a very tasty and tempting sports car your insurance rates go up, etc. Similarly, interventionist military action, if we accept your premise, also tends to increase your military costs.

All of this presupposes that our strong military and interventionism was the cause of the terrorist attacks. I don't believe that, and see damn little evidence that it was.

Posted by: Dean Esmay on August 6, 2002 02:27 AM

How are military expenditures a form of insurance when that term is defined as follows:

1 a : the business of insuring persons or property b : coverage by contract whereby one party undertakes to indemnify or guarantee another against loss by a specified contingency or peril c : the sum for which something is insured?

Military expenditures do not compensate a nation for injuries caused by the aggression of another nation or terrorist group. And insurance cannot deter a prospective injury from occurring.

I think the metaphor is a singularly inapt one.

Posted by: Herr K. on August 6, 2002 12:56 PM

OK, Herr K., then a military is like bars on your windows and a Rottweiler in the yard. And that is very much an expense that one would rather not have, albeit one that in the real world is often unavoidable. The opportunity cost of a strong military is very high, but the consequences of not having one (at least for a country with any pretensions to global influence) are too awful to coutenance.

Posted by: David Gillies on August 6, 2002 01:57 PM

Military expenditures are not a form of insurance per se - but (if you can remember not to use them to _start_ trouble), they are a protection against greater loss. Another example is Quality Assurance. It's a non-value added activity, but if you don't do it sufficiently, something bad is bound to happen...

Posted by: MarkM on August 7, 2002 12:15 PM


Isn't that Al's point above, ie, that to the extent the economic question concerning the war isn't preposterous, the relevant question is "What is the cost of not prosecuting the war?" and not "what's the cost of prosecuting it?"

Posted by: Herr K. on August 7, 2002 01:04 PM

Herr. K, I think you're correct that the relevant question is the likely outcome of not having adequate defences rather than the cost of having them. I was actually agreeing with you :).

All the analogies we've come up with are lacking, of course - they ignore the force projection capability aspect of having a military. Unless you let your Rottweiler out to attack little old ladies.

I think the root of the whole 'war is good for the economy' trope is based on the old fallacy (which fortunately no-one has been silly enough to air) that breaking windows isn't so bad because at least it drums up trade for glaziers. Sure, Bell Helicopter made out like bandits during Vietnam, but I think on the whole the war was bad for the US economy...

Posted by: David Gillies on August 7, 2002 06:35 PM


The "broken-window" fallacy is from Bastiat's classic essay "That Which Is Seen and That Which is Unseen") (or something to that effect, my French sucks, I grew up in Brooklyn), upon which Henry Hazlitt based his book "Economics in One Lesson."

I think military expenditures are inefficient in a way significantly different, in quality and degree, than destroying property (which destroys wealth) to create work (cf, the argument hurricanes are good because they put people to work...). Military expenditures redirect capital from the private sector--away from the uses to which individuals would ordinarily devote it (and which are usually more profitable (ie., wealth-producing) than the public sector functions to which it is then devoted--but it does not destroy and thereby reduce aggregate capital and society's wealth. (No doubt there a people at the Cato Institute that would disagree.)

In a word military expenditures (because inefficient) may reduce future capital but not destroy current capital.

Posted by: Herr K. on August 8, 2002 02:06 PM