December 05, 2004

The Iraq War as Present Value

by Demosophist

Richard Posner, on the brand-spanking-new Becker-Posner blog, manages to say in a few succinct words what I've been attempting to say in a far less economical way for over a year. One implication of the exposte anti-war argument concerning Iraq is that the Administration knew the probability of Saddam having WMD was zero, but chose to invade anyway for some unambiguously self-serving reason. That's essentially what all the "Bush lied" talk was about. The other, less well articulated, implication is that there was no estimate made of the probability, because it didn't matter. We would have attacked whether the probability of a future WMD attack from Saddam was 1 or 0, or anything in between. The former implication, that we knew the probability was zero and acted as though it were closer to 1 really has little merit. Had we known with certainty that there was no threat, then there'd have been no debate at all about the evidence. If it had been faked, the fakery would have been undetectable, because the deception would have been coldly premeditated. There would have been no bungling attempt at a poorly constructed "yellow-cake" document from Niger. But the second implication at least has some small degree of merit. The administration may well have decided that the real issue was not whether Saddam was about to attack the US, but whether in the fullness of time, a radicalizing Arab Middle East would have become an unacceptable threat. If an administration had concluded that this was a significant probability, then the only question left would be where to intervene.

Which Middle Eastern nation presented the ripest opportunity? With the objective of "interfering with" the gradual development of a totalitarian Middle East, not only potentially capable of attack, but also capable of withholding oil resources (though there's an inherent conflict between obtaining resources to build weapons and cutting off the inflow of petro-dollars) Iraq seems ideal. Not only did it have a recalcitrant leader with parochially flawed strategic judgment, but he was also a crazy murdering S.O.B. About as unlovable a character as they make.

And geographically, Iraq is at the center of things in the Middle East. From there we could not only launch "our vanguard" for liberal democracy, to counter the Salafist vanguard for militant Islamism, but if necessary we could have a launch point for operations against other Middle Eastern threats as they emerged.

By the numbers, from Posner:

Suppose there is a probability of .5 that the adversary will attack at some future time, when he has completed a military build up, that the attack will, if resisted with only the victim's current strength, inflict a cost on the victim of 100, so that the expected cost of the attack is 50 (100 x .5), but that the expected cost can be reduced to 20 if the victim incurs additional defense costs of 15. Suppose further that at an additional cost of only 5, the victim can by a preventive strike today eliminate all possibility of the future attack. Since 5 is less than 35 (the sum of injury and defensive costs if the future enemy attack is not prevented), the preventive war is cost-justified.

A historical example that illustrates this analysis is the Nazi reoccupation of the Rhineland area of Germany in 1936, an area that had been demilitarized by the Treaty of Versailles. Had France and Great Britain responded to this treaty violation by invading Germany, in all likelihood Hitler would have been overthrown and World War II averted. (It is unlikely that Japan would have attacked the United States and Great Britain in 1941 had it not thought that Germany would be victorious.) The benefits of preventive war would in that instance have greatly exceeded the costs.

In the case of the Iraq War the probability isn't really made concerning Saddam, because the larger probability concerns the Islamist movement, as a whole. Having watched the Middle East slowly radicalize over a period of 40-plus years, leading inexorably to the 9-11 attacks... whether Saddam were an imminent or a growing and gathering threat may simply not have mattered.

Consider this analogy: You have reason to believe, with a probability of 0.5, that there's a ticking time bomb in a room of a larger building held by a terrorist gang, and the bomb is capable of destroying the better part of a city that can't be evacuated. (You can't evacuate the world, yet. Nor could we even evacuate a single country.) Your decision is whether to forcefully enter the building, which could endanger some hostages. The fact that the room may also contain fully automatic weapons and other ordnance wielded by thugs is important to your plan of entry, but it matters little to your decision of whether or not to enter if your cost-benefit calculation dictates that you must attempt to defuse the bomb. (I know there are a few problems with this analogy, such as the possibility that the thugs might trigger the bomb themselves if you enter, but in this scenario we assume that for some reason they can't do that.)

I guess the question arises as to whether you consult the public, but lets suppose that the calculations about the bomb involve some specialized knowledge that can't be shared with the public. Not only that, but you don't happen to enjoy a great deal of favor with the public so that they cannot simply take your word for the validity of your prediction. Your assessment of the probability of the larger risk is 0.5, but even after being told of it theirs is more likely to be something like 0.05 or even 0.005.

So, instead of presenting the larger but less familiar threat you present the lesser, but more familiar one... and hope that you can make a good enough case to convince people that the action is necessary.

If one consults the history of the run up to the Iraq War, the Bush administration did attempt to make a case that was concerned with the introduction of liberal government to Iraq, at least as a kind of humanitarian mission coupled to a larger strategic security. This argument sufficiently impressed Paul Berman and Bernard Kouchner that they each wrote eloquent defenses of the pending action on those liberal grounds alone. Neither, however, were widely read. Nor did they generate much of a political groundswell. One must doubt that had the case been made to the public on these grounds it would have had much heft. The lack of interest may simply have been because the case seemed more humanitarian than pragmatic. Even though there is a profoundly pragmatic reason for humanitarian action one is liable to be skeptical of such "unrealistic" naiveté'.

At any rate, the decision calculus of the public is probably a different topic, and in that calculus the issue of Saddamite WMD was the top priority. However, it's important to recognize that a top priority doesn't mean it was the only priority that concerned Americans, though it had been virtually the only priority of either "Old Europe" or the UN itself.

I guess the only point left to make is that if you're convinced of the growing threat of Islamo/fascism as even greater than the short term threat posed by Saddam, then you're also likely to be convinced that there's a greater expected cost to Arab Middle Eastern populations for the "act later" scenario. (See Armed Liberal for that argument.) In terms of Posner's calculus, acting in Europe during 1936 would not merely have saved more allied, but also a great many more German lives. (Few would have been cynical enough to have predicted the deaths of six million Jews .)

(Cross-posted by Demosophist to Demosophia and Anticipatory Retaliation)

Posted by: Demosophist at 11:50 PM | Comments (9) | Add Comment
Post contains 1309 words, total size 8 kb.

1 I don't think you can necessarily use "what ifs" to justify war after-the-fact. In fact, it sounds more akin to trying to cover one's ass to suggest other reasons for invasion when the primary reason turns up fruitless. The Bush administration may not have "lied" per se, but when you have Condi, Cheney, Rummy, and Bush going before committee after committee and ally after ally (not to mention the UN) with their documented "proof" of quantities, locations, and types of WMDs...well, you have to wonder who lied to who. Did they lie to the above, or did someone lie to them (and, if that's the case, why and how - after the tragedies of September 11th - could information-gathering be so incompetent)? Either way, it doesn't leave one being very confident of this administration.

Posted by: Venom at December 06, 2004 09:36 AM (dbxVM)

2 I don't think you can necessarily use "what ifs" to justify war after-the-fact. In fact, it sounds more akin to trying to cover one's ass to suggest other reasons for invasion when the primary reason turns up fruitless. I don't understand. In what sense am I using "what ifs" to justify the war after the fact? I'm simply using, as has Posner, a rational calculation of expected value to suggest that waiting until the probability of attack is 1 is a synthetic and irresponsible standard. Do you understand the concept being used here? Expected value involves a combination of the costs of certain resultants combined with their probability of occurrence. It's a decision methodology intended for use ex-ante. However, looking at how it could have been used ex-poste is a way to justify the use of the decision method in the future, not the past war itself. The Bush administration may not have "lied" per se, but when you have Condi, Cheney, Rummy, and Bush going before committee after committee and ally after ally (not to mention the UN) with their documented "proof" of quantities, locations, and types of WMDs...well, you have to wonder who lied to who. The error in your thinking is to assume that "proof" implies certainty, and that certainty is required in order to act. Both assumptions are wrong, which is what the "expected value" argument is attempting to clarify. In other words, it is unacceptably expensive to use either a "wait until attacked" or some other ad hoc decision rule. Did they lie to the above, or did someone lie to them (and, if that's the case, why and how - after the tragedies of September 11th - could information-gathering be so incompetent)? Either way, it doesn't leave one being very confident of this administration. You have a valid point here. If I might tease it out a little... In the case of the soviets we had a fairly good idea of what the probabilities actually were, associated with a particular venture, because we had excellent intelligence. If you don't know whether the probability of attack is 0.5 or 0.05 then you have a problem, because you have to assume the higher number for the sake of security (especially if the consequences of being too cautious are catastrophic). And that means the odds of making a wrong, or too-costly, decision is higher. In retrospect we now know who was actually lying about Saddam having WMD. It was Saddam, who thought it a tactical advantage with respect to the Iranians that they think he probably had them. He wasn't trying to deceive us, but his closest geographical and historical enemy. His parochially flawed judgment was a bigger problem than we thought. But that doesn't really change my argument, because I don't think the decision to attack was really based on whether he actually had WMD. Had we known with greater certainty, however, we might have been able to adjust the rationale for the war... or (ideally) have found some way for him to make the first aggressive move (such as to shoot down a UN U2, which he had threatened to do). We would have taken a different, and presumably less costly, route to the solution.

Posted by: Demosophist at December 06, 2004 10:08 AM (7AGFb)

3 "I don't understand. In what sense am I using "what ifs" to justify the war after the fact? I'm simply using, as has Posner, a rational calculation of expected value to suggest that waiting until the probability of attack is 1 is a synthetic and irresponsible standard. Do you understand the concept being used here?" Actually, I do. But I think you could make the case that any country's future probability for invading the US is 1, given a long enough period of time. In any case, I think it's highly unlikely that the Bush administration used this "highly scientific" approach to evaluate the threat of Iraq, considering that if they had, the threat from other nations such as North Korea probably had (and continue to have) a greater probability of attacking the US. "The error in your thinking is to assume that "proof" implies certainty, and that certainty is required in order to act." No, that's true, "certainty" isn't necessarily "required" to act, but having it definitely helps when you're sending people's sons and daughters into harm's way. And when they're making the case all over the world about Iraqi WMDs, they're basically giving the impression they know what they're talking about. Why put in so much effort around the world to bolster their case, only to then say afterwards "well, we never said for certain that WMDs are there." C'mon, even you can see that they were aiming for unequivocable concrete evidence that Iraq had WMDs. Proof or certainty, they basically told the world that they existed, and they made their case that these WMDs warranted an invasion.

Posted by: Venom at December 06, 2004 11:02 AM (dbxVM)

4 I think I am getting light headed ... must keep it togeather. All joking aside I can say that a majority of the people who disagree with the Iraq Invasion argue the fact that Afghanistan was allowed with no arguements against the administration, even all out support. Even though I dont agree with the administration on Iraq and contend that it was a botched abortion, I still support the Armed Forces in accomplishing their mission. They have accomplished their mission to date with hard work and valor, regardless how empty the reasons were behind the orders given to the Armed Forces, they have followed them to the "T". Its a lame arguement to attack anyone who disagrees with the administration as "not supporting the troops", a hollow arguement akin to wanting to excommunicate parishiners for questioning the faith of priests who participated in sexual misdoings.

Posted by: Salamander at December 06, 2004 11:49 AM (F26eZ)

5 Even though I dont agree with the administration on Iraq and contend that it was a botched abortion, I still support the Armed Forces in accomplishing their mission. They have accomplished their mission to date with hard work and valor, regardless how empty the reasons were behind the orders given to the Armed Forces, they have followed them to the "T". Well, that's an important distinction. The troops almost certainly appreciate your support, even though you may not agree with their mission. Its a lame arguement to attack anyone who disagrees with the administration as "not supporting the troops", a hollow arguement akin to wanting to excommunicate parishiners for questioning the faith of priests who participated in sexual misdoings. Well, at least it's an argument I'm not making. I'm not so sanguine about the others who might disagree with the Administration, especially those like Ted Rall and Michael Moore who tend to characterize the people who constructed the Fallujah slaughterhouses as "freedom fighters." I tend to think those folks aren't supporting the troops, and in fact are probably traitors.

Posted by: Demosophist at December 06, 2004 12:09 PM (7AGFb)

6 Do any businesses use present value calculations for thigns this big? In school they taught us to also take into account risk (variance).

Posted by: actus at December 06, 2004 09:19 PM (YxF4W)

7 The Law and Economics movement often looks as if it were a whole lot of thinking based a few completely uncited factual assumptions. Like this one and the .5 chance of attack.

Posted by: actus at December 07, 2004 08:25 AM (YxF4W)

8 Actus: I don't think they expect you to buy their 0.5 estimate. What they're proposing is a certain clarity about the decision process, primarily because many of the arguments against the war simply omit much of that process. (For instance, they assume you should only act on certain knowledge.) o any businesses use present value calculations for thigns this big? In school they taught us to also take into account risk (variance). Well variances (betas) are simply estimates of the level of certainty or uncertainty combine with estimate of monetary benefits or costs. Essentially this is what risk is all about. So, yes, businesses use precisely this method... the only difference being that they apply a specific timeline for a project, with a discount rate that is supposedly produced by a market that incorporates all of these risk calculations into the rate. Since we don't have a market to tell us the rate, however, we have to use formal methods to derive it.

Posted by: Demosophist at December 07, 2004 02:05 PM (7AGFb)

9 'Well variances (betas) are simply estimates of the level of certainty or uncertainty combine with estimate of monetary benefits or costs. Essentially this is what risk is all about. So, yes, businesses use precisely this method' Really? they're precisely risk neutral when it comes to things like war and the middle east? surprising.

Posted by: actus at December 07, 2004 04:53 PM (YxF4W)

Hide Comments | Add Comment

Comments are disabled. Post is locked.
34kb generated in CPU 0.018, elapsed 0.1584 seconds.
119 queries taking 0.1473 seconds, 258 records returned.
Powered by Minx 1.1.6c-pink.