C. Causation (cont'd.)

    2. Redundant causation

    Kingston (p) v. Chicago & N. W. Ry. Co. (d) 211 N.W. 913 (1927)

1. Facts of the case?
2. You are the judge: There is one fire, resulting from RR sparks. Would you award damages?
3. You are the judge: There were two fires, one from the RR sparks, another from lightning. Either would have destroyed plaintiff's property. Would you award damages?
4. You are the judge: There were two fires, both resulting from negligence, either sufficient to destroy plaintiff's property. Plaintiff sues only one of the potential defendants. Would you award damages?
5. What marginal damage did defendant's fire cause?
6. Can the plaintiff rely on the "but for" test?
7. Why liable even so?
8. P: Did the appeals court hold defendant liable? yes

    Illustrates redundant causation

    2 competing doctrines here:
    doctrine of cause-in-fact = the "but for" test: places the burden of proof on the plaintiff to show that the damage would not have without the defendant's action
    this case fails that test

    doctrine of multiple sufficient causes: defendants must prove that at least one cause was not negligent
    this prevents multiple tortfeasors from escaping liability

Posner and Friedman disagree about this case.
Both tell a hunting story. For Friedman:
    Al and Bill both shoot Carl

    Posner: "it would be an economic mistake to let both off scot-free." (P184,6e)
    Friedman: "[I]n clear cases of redundant causation, where there is no risk of conspiracy, [tortfeasors] ought not to be held liable...." (F194)
    Rationale: MC of the 2d injurer = 0
    As a practical matter, Friedman recognizes that this risks perverse incentive effects by encouraging "redundant" causation to escape liability (F)

    In this case, I side with Posner
    Outcome: ex post MC of the 2d injurer = 0
    Efficiency: provide the right ex ante incentive

    Consider n independent, sufficient causes with = probability: p
    Total probability of avoiding injury = (1-p)ⁿ
    => marginal probability of injury (MPr) = (1-p)^(n-1) - (1-p)ⁿ

    MPr falls as n rises because of rising opportunities for redundant causation

    Friedman's Ex:
    probability of either hunter shooting Carl = 10%
    Value of Carl = $1M

P(Carl safe) MP(Carl shot) 1 hunter 90% 10% => Exp. MC = $100K 2 hunters .9 x .9 = 81% 9% => Exp. MC = $90K     Per Friedman text:     Efficient penalty =>     MChunter 2 = .09 x $1M for 1 hunter shooting Carl         + .01 x $0 if both shoot Carl         = $90K     Problem: Friedman doesn't consider benefits, only costs

    Suppose the MB of hunting for each hunter is $91K
    Then with 2 hunters: TB=$182K; TC=$190K => the hunting is negligent
    => optimum = 0
    General Pigouvian rule: set price = MC at the optimal Q.
    => always pay if you shoot Carl and every injurer is negligent

the general principle used in the common law probably gets the right outcome more often than not.

    Notice Kingston v. Chicago relied on this distinction--Chicago Ry is liable if both fires are negligent
    In Cook v. Minneapolis, one fire was natural, so the negligent party was not liable
    You are not liable if you toss a burning cigarette into a house hit by lightning and on fire.