js5389's version from 2015-12-10 20:31
|Five Main Collective Action Problems||(1) Prisoner’s Dilemma|
(2) Insufficient Value
(3) Disparity in Resources
Impact of the Remedy
Certainty of Termination
|Hansberry v. Lee Facts||In H v. L, black family that had just bought a house in a neighborhood governed by a racially restrictive covenant; white homeowners sued to have the covenant enforced, and said that matter had already been decided in Burke v. Kleinman, in which the π purported to rep “all homeowners.”|
|Hansberry v. Lee Holding||Black homeowners were not adequately represented in the first case|
|Hansberry v. Lee Takeaway||Ct basically trying to find a way that racist whites don’t represent blacks. Homeowner class in Burke had been certified in equity (i.e. informally) so there was not enough procudural rigor to ensure fairness and adequacy of representation to the other class members (“good law of Hansberry”).|
|Mullane v. Central Hanover Bank Facts||Mullane = lawyer for income beneficiaries of a trust in NY. Amounts are very small but the law says the banks need to get approval of beneficiaries to close out the accts ever yr. Mullane is a de facto class rep and thus the state has created a class thru the statute requiring him to rep bens.|
|Mullane v. Central Hanover Bank Holding||all named beneficiaries need mailed individual notice; this would suffice for the class|
|Mullane v. Central Hanover Bank Takeaway||The principle behind the Mullane holding (reasonable notice reasonably intended under the circumstances is required to functionally protect your rights we’re uncomfortable w/having someone else vindicate your rights, so we give absent parties a mechanism of representation) was a good one, but it was too formalistically applied on the facts of the case – this just ended up being really inefficient and unnecessary b/c no one cares about $5 in a rando NY bank account, and ended up costing the beneficiaries $, b/c the transactional fees that M gets come out of their mini-trusts. The idea of notice is nitce, but these people have NO incentive to effectuate their rights in any case. Example of the importance of protecting a party’s chose in action (interest in being able to go to ct)|
|Holland v. Steele Facts||Inmates at Dade County jail are not allowed access to lawyers, so Holland seeks to certify a class consisting of all present and future jail inmates to file a civil rights 23(b)(2) class action against the jail.|
|Holland v. Steele Holding||Class can be certified hinged a lot on adequacy of rep (law firm often rep’d class actions), although other three prereqs (numerosity – 40 π’s now, unknown in the future; commonality – yes all in same situation; typicality – yes). No notice requirement b/c its (b)(2) [stop not letting us call lawyers!]|
|Holland v. Steele Takeaway||“this is a dangerous case” b/c it was clearly a lawyer-driven test case. The Ms. Chu problem – you become a member of this class when you get arrested in Dade County and tossed in jail, and you have no chance to opt out or even know about it beforehand. Jail opposes it b/c it would be cheaper to settle out with Holland, and its not in Holland’s interests b/c a) his lawyers aren’t paying him person attention, they’re dealing with a class and he’s just one member, and b) in term of preclusion, if H won, jail would be precluded against a 2nd π the same way they would be if H’s class won.|
|Rhone Poulenc Facts||case gets to appeals ct on a writ of mandamus (23(f) appeal to review an order granting or denying a class action). Π’s wanted to get certified as a class b/c they are all hemophiliacs who allege they got HIV from R-P during blood transfusions from untested blood; want to be a class so they don’t have to be individually identified (stigma against ppl with HIV). Just wanted to certify the class to determine whether ∆ was negligent, then would bring each case individually (“serendipity theory” – w/o harm A, harm B couldn’t have occurred).|
|Rhone Poulenc Holding||decertifies class b/c it would be inefficient and prejudicial to ∆’s (re-examination problem: every π would be bringing in the same evidence). Says π’s “serendipity theory” [try ∆ for negligence (general causation), then try individuals (specific causation)] shouldn’t be tried for the first time as a class action, and that the risk to the ∆ would be inordinately high, since 12/13 cases have gone in favor of ∆’s, but π’s could win on a one-shot trial, which would be unfair b/c it could ruin the company by forcing them to make massive payments|
|Rhone Poulenc Rationale||- Inefficient – each π would end up having his own case anyway, so there would be no efficiency gain from class cert.|
- Re-examination problem – the jury in all of these cases would essentially be assessing the findings of the jury in the class action case determining ∆’s negligence (Note: Posner is reading the 7th amendment wrong here [we actually don’t want juries on appeal assessing matters of fact determined by trial juries] but his point here is right this would just be super inefficient)
|Rhone Poulenc Takeaway||- Issacharoff: Posner also looked at the fact that ∆’s had won 12/13 cases as evidence that they probably weren’t negligent anyway. This is an erroneous conclusion – maybe π’s were just trying out other legal theories – not for sure that the “tail end” of the class action amplification effect really would be disproportionate to the π’s win rate. These cases were not random draws, and the 12/13 is not indicative of ∆’s likely guilt necessarily.|
- Note: pre R-P, there was a settlement offer on the table for $1.2 mill; after the class got decertified, they settled as a class anyway for $700 mill. So, “this is an $800 mill decision” – shows huge impact!
- Castano case: you have to be able to determine if you can carve at the joints. Ie can you get a clean separation of issues from the first case to the subsequent application. IF you can’t, you’ll have no efficiency gains.
|Jenkins v. Raymark||Judge Parker; court decided to use a 23(c)(4) class to determine whether a state of the art technology defense was available to ∆ companies and if punitive damages should be available; did not turn on π’s at all|
|Cimino||SI tries to save the day, gets pwnd. Tried to “replicate” how the litigation would play out by considering a variety of factors, allow juries to hear 10 π’s who represented the various types of people with various types of symptoms and possible contributing factor combinations, and basically certify a class of all pending asbestos litigation claims in that district.|
|Amchem v. Windsor Facts||23(b)(3) attempt to resolve these cases once and for all. ∆’s agree that no one will settle with any one π unless the π’s as a whole agree to settle with the all the ∆’s. Now ∆’s feel they have some leverage in that they’ve solved their collective action problem. The ∆’s want to settle the entire game. Π’s are like aight.|
|Amchem v. Windsor Holding||RBG decertifies class! There is no structural assurance of fairness if you’re going to bind people in absentia and prospectively [Hansberry!] – there were basically 3 subgroups of π’s (sick, may become sick, and not brought claims yet), and there was no way of preventing the first group from getting all the $ and no one to rep the last group.|
- Hinged a lot on adequacy of representation and commonality [3 different subclasses have 3 different desires ITO settlement amount and timing]
- RBG also argued that π’s had no leverage b/c they were willing to settle, so the amount they were getting was artificially low. SI SAYS THIS IS WRONG – π’s can go to trial at any time (that’s leverage!)
- She also says that manageability is not as important as predominance requirements ITO a settlement class b/c you’re not going to trial anyway!
|Amchem v. Windsor Takeaway||Two takeaways from Amchem: 1) Adequacy of representation is key in crafting a class! And 2) we can finagle rule 23 to make predominance the main concern in settlement cases.|
|Ortiz v. Fibreboard||right after Amchem, parties agreed to a 23(b)(3) class action and the court decertified: “court took the mass action out of the public domain” – every single major asbestos producer has gone into bankruptcy and 1000’s of people have died b/c courts refuse to sacrifice procedure for efficiency.|
|Martin v. Wilkes Facts||City of Birmingham had in place a “make whole remedy” consent agreement with black firefighters that resulted from earlier litigation with a group of black firefighters. City said forgo back pay and we’ll bump you up in promotion order (this was good for city [saved $5 mill] but screwed over future while firefighters). This case comes from whites suing for not getting promotions b/c less senior blacks have jumped the line and gone ahead of them. Q: Should the earlier consent agreement bind the whites?|
|Martin v. Wilkes Holding||If current parties want outside parties to be bound by their case, it is up to those existing parties to join them! (Rule 20)|
|Martin v. Wilkes Takeaway||- Underlying policy rationale: You can’t make a party intervene (Rule 24 is permissive!) – the parties actually in the case (black ff’s and the city) are in the best position to know who they’ll want to preclude from judgment later on, so its up to them to join those people!|
* We must put the burden of preclusion on the parties to the litigation!
* Also rep became inadequate once blacks agreed to settle — whites are adversarial w/ city, while blacks are now friendly – how can rep be adequate?