Summary
finding that "[t]he settlement agreement would violate the [FLSA] because it "is unconscionable to try to take away the FLSA rights of all workers, whether or not they choose to join in affirmatively"
Summary of this case from Woods v. Vector Marketing CorporationOpinion
No. C 06-06493 WHA.
June 19, 2007
ORDER DENYING PRELIMINARY APPROVAL OF PROPOSED CLASS SETTLEMENT
INTRODUCTION
This order DENIES a joint motion to preliminarily approve a proposed settlement. To put this case back on track, the earlier motion for class certification (temporarily taken off-calendar to consider the settlement motion) shall be heard on AUGUST 2, 2007, AT 8:00 A.M. The case schedule otherwise remains as stated in the case management order dated January 11, 2007. This denial is without prejudice to further attempts to settle this case, but we will not delay our case schedule again to accommodate further settlement efforts. Any settlement after the formal ruling on the certification motion will need to be consistent with that ruling as well as the instant order.
STATEMENT
This action began as a wage-and-hour case alleging an FLSA collective opt-in action and a California class of former workers under California overtime laws. Three former workers sued Oracle Corporation for unpaid overtime due to alleged misclassification of "sales consultants" as exempt employees. Recently and without moving to amend the complaint, counsel announced that they had reached a settlement, much expanded in coverage from the original.
Now, a nationwide class involving ten or so job classifications is in play. Under the settlement, all wage-and-hour rights (not just overtime) of putative class members would be completely extinguished and replaced by an exclusive claims procedure. By expressly obligating itself only on a "claims-made" approach, Oracle would pay only those who submit claims up to a total of nine million dollars less all fees and expenses. Counsel wants $2.25 million in attorney's fees and $75,000 in expenses. In addition to their own shares of the settlement, $45,000 total would be paid to the three named plaintiffs as "incentive payments." Costs of administration would also be deducted. Because it will be a "claims-made" settlement, there will be no residue. All unclaimed amounts will revert to Oracle. Counsel now desire preliminary approval under Rule 23(e) and recommend notice be sent by mail to last known addresses of 1500 or so workers granting them a brief period for filing claims — after which all of their claims and rights would be forever barred, even as to those who never receive actual notice or submit a claim.
ANALYSIS
Class actions are ideally suited to the efficient resolution of numerous parallel claims in a single proceeding and to encourage the pooling of small claims against a common target. They are an engine of justice in our federal courts. But they can also lend themselves to abuse. One form of abuse is collusive settlement. Once the named parties reach a settlement in a purported class action, they are always solidly in favor of their own proposal. There is no advocate to critique the proposal on behalf of absent class members. That is one reason that Rule 23(e) insists that the district court vet all class settlements. While always giving deference to counsel's views of the advisability of a settlement, a district court may not simply rubber stamp stipulated settlements. We must be careful to make sure absent class members will be treated fairly. Staton v. Boeing Co., 327 F.3d 938, 959-60 (9th Cir. 2003) (assess for "actual fraud, overreaching or collusion").No class has yet been certified in this action. No named plaintiff has yet been certified to speak for any class. No counsel has yet been designated to represent any class. Issues of typicality, commonality, adequacy, manageability, and the possible reach of a class outside of California have not yet been decided. The proposed settlement, therefore, comes at a time before any representative or counsel has been appointed to lead any class. The complaint stands unamended to conform to the dramatic expansion of claims and persons affected by the settlement.
* * *
As the Court reads the Settlement Agreement, it would "extinguish" any and all claims of the nationwide worker class whether or not they receive actual notice and whether or not they submit claims, with all unclaimed amounts reverting to Oracle. In response to the Court's questions, counsel tried to put a softer spin on the release and to say this was not the intent. But we must be guided by the actual written agreement and release, for that is the instrument that will be interposed by Oracle in various other courts across the country to bar workers' claims. Because the written instrument simply does not comport with counsel's softer spin, this order will take the time to walk through its interlocking provisions to show the draconian scope.
Paragraph 14 provides that "[e]ach Eligible Class Member shall be deemed to have jointly and severally released and forever discharged [Oracle] from any and all `Settled Claims,' whether known or unknown." Paragraph 15(b) provides "this Agreement is intended by the Parties to be a release and covenant not to sue as to the Eligible Class Members which extinguishes all Settled Claims and precludes any attempt by the Eligible Class Members to file or continue prosecution of a lawsuit or otherwise pursue an individual claim for relief with any local, state or federal agency or court against [Oracle] with respect to the Settled Claims."
In turn, "Settled Claims" means "any and all claims that were asserted or could have been asserted in the Complaint [in this action], including various California Labor Code Sections and Business Professions Code Sections" (¶ 2(s)) (emphasis added). This definition further expressly states: " In addition, Oracle contends that this settlement will act to release claims under the other state statutes, regulations and orders listed in Exhibit B of this Settlement Agreement under which such Settled Claims may have arisen." Exhibit B is a list of 150 labor statutes in 35 states. What "could have been asserted" is a question of interpretation. In settlement practice, however, the "could have been" language is used to obtain the broadest release possible. In its broadest reach, Oracle's gloss is plausible, i.e., all further state laws would be extinguished as they "could have been" asserted by way of amendment.
We must remember that the FLSA was expressly asserted in the complaint, so the release purports to extinguish all FLSA rights. In this regard, the proposed notice flat-out states (page 19): "Eligible Class Members waive and release their right to sue with respect to any Claim under the Fair Labor Standards Act of 1938." Contrary to the spin, the FLSA release is not limited to those who opt-in but extends to all "Eligible Class Members."
"Eligible Class Member" is a defined term meaning "all persons who are members of the Settlement Class except for those persons who have validly requested exclusion from the settlement as provided in this Settlement Agreement and the Class Notice" (¶ 2(l)). In turn, "Settlement Class," yet another defined term, means "all persons employed by [Oracle] in one or more covered Job Codes and organizations listed in Exhibit E from the commencement date set forth therein [starting in 2002] through the date of the Preliminary Approval Order," estimated at 1485 individuals (¶ 2(v)). Exhibit E includes about ten or so different jobs. So, Eligible Class Members includes all but the opt-outs. It is not an opt-in class, as stated by counsel. It is everyone but those affirmatively excluding themselves.
The Settlement Agreement provides that even those never receiving notice will be bound by its terms. This feat is accomplished by stipulating that the method of notice will be deemed "the best practicable notice." The notice method would use "the most current mailing address information available to" Oracle. Undelivered envelopes returned with forwarding addresses would be sent to the forwarding address. If no forwarding address is provided, then the notice will be resent "after the address is updated through a computer search," all according to Paragraph 20, which then provides (¶ 20(d)) (emphasis added):
In the event the procedures in this Paragraph 20 are followed, Defendant shall be deemed to have satisfied its obligation to provide Class Notice to a member of the Settlement Class, and if an intended recipient does not receive the Class Notice, the intended recipient shall nevertheless be bound by all terms of the Settlement Agreement and the Final Order.
Significantly, this means that those not receiving notice or opting in will nonetheless have their claims extinguished. This is repeated again in Paragraph 23 — those who do not submit a claim — even if because they did not get notice — shall be nonetheless be bound by the release (¶ 23(e)):
Although Eligible Class Members who do not submit a valid and timely claim form will not receive a Settlement Award, such persons shall nonetheless be bound by all the terms of the Settlement Agreement and the Final Order and Dismissal with Prejudice.
To close the loop, Oracle would get to keep all funds (including interest) attributable to those who fail to file forms even if they never receive notice (¶ 27(d)):
Reversion to Defendant. The Claims Administrator will pay to Defendant any unclaimed settlement funds and the interest on those unclaimed settlement funds. Unclaimed settlement funds include (i) the amount of all Settlement Awards that would have been made to those who timely opted out of the settlement; (ii) the amount of the un-cashed settlement checks issued to those Eligible Class Members who submitted Claim Forms but did not cash their settlement checks within 180 days of the date they were issued; and (iii) the amount of Settlement Awards that would have been given to those who failed to timely submit a Claim Form but did not opt out of the settlement.
The Court is at a loss to understand how counsel can portray the release as benignly as they have. The proposal — as signed — is unfair to absent class members for the following reasons.
1. THERE IS NO FINAL AGREEMENT AND THIS MOTION IS PREMATURE.
Amazingly, in the process of explaining the proposal to the Court, a significant disagreement between the two sides emerged, namely, the parties disagree fundamentally over the scope of the release to be extracted from the class. They agree that the release would forfeit "any and all claims that were asserted or could have been asserted" in this action. The parties have left open, however, a glaring disagreement as to what this language means. In addition to all federal claims, Oracle expressly contends that this settlement will act to release overtime claims under 150 laws and regulations in 35 states. As stated, the Settlement Agreement flat-out provides: " In addition, Oracle contends that this settlement will act to release claims under the other state statutes, regulations and orders listed in Exhibit B of this Settlement Agreement under which such Settled Claims may have arisen" (¶ 2(s)). Plaintiffs do not agree.
This is such an important point — affecting over seventy percent of the proposed class members who reside outside California — that it cannot go unresolved and left to chance. Class members are entitled to a notice that would advise them of the true effect of the proposal. It is counter-productive merely to tell them that the sides disagree over what they would give up and force them to figure it out. As a matter of prudence, moreover, we should not impose the uncertainty of this major question on absent class members and ultimately on various judges around the country, inviting a donnybrook of inconsistent outcomes. Most of all — at the most fundamental level of contract formation — there is no final offer and acceptance, given the dispute over this material term. This is dispositive. There simply is no final agreement on all material terms ready to present to any proposed class.
It is true that at the hearing on this proposal, counsel tried to mute their differences. In the end, however, the Court must evaluate the operative agreements and notices, for these are the documents that Oracle would interpose to bar subsequent workers' suits. The Settlement Agreement and its "In addition, Oracle contends . . ." clause is unacceptable. At this juncture, simply removing the clause would not solve the problem. An express agreement on this point of contention would be necessary.
Rather than stopping here, however, this order will go ahead and address a number of other problems with the proposal. In doing so, this order will assume that Oracle's position as to the scope of the release would prevail. This presumption is adopted out of caution and concern for absent class members against whom Oracle would assert the release.
2. THE PROPOSAL WOULD NOT BE FAIR TO ABSENT CLASS MEMBERS.
Throughout the United States, all affected workers' rights to recover all unpaid wages under state or federal law or regulation would be completely extinguished by the proposal. This would be true whether or not a worker ultimately files a settlement claim or whether or not the notice letter is even mailed to a wrong address or whether or not the worker even receives actual notice of the proposed settlement. In place of these rights, the proposal would substitute a fleeting opportunity to file a claim for money. Workers who fail to receive the notice (due to changes in address or other delivery problems) or who put it aside unread due to the press of other matters or who simply do not prepare and file a claim on the tight timetable in the notice would lose all rights. Even those who might trouble themselves to file a claim form would not receive all that they deserve, if the operative complaint is accepted, but only a fraction thereof.
The proposed contemplates preliminary approval on June 21, mailing of notice on July 6, with the last date to object, opt-out or to submit a claim being August 20, even for those who receive no letter at all or receive it late due to delivery problems and address changes. This rushed timetable is curable, but the current compressed schedule is too short.
Such a scheme would be a bonanza for the company. With a single stroke, the company would wipe the slate clean of all its wage-and-hour liabilities for all affected workers nationwide, not just for overtime claims in California. Oracle would only have to pay some workers, i.e., those who actually submit a timely claim — and, even as to them, it would only have to pay a fraction of what the complaint alleges is really due. Oracle's liability would be limited on a "claims-made" basis, as Paragraph 33 of the Settlement Agreement provides. But the release is not imposed on a "claims-made" basis. Far from it. All workers' rights nationwide would be obliterated even if only a few submit claims.
Paragraph 33 states: "The parties agree that this Settlement Agreement does not create or establish a settlement or common fund, and that all claims for payment shall be submitted directly to and paid by the Claims Administrator. The parties represent that the Funds Available for Settlement is a settlement amount that takes into account the probability that some Eligible Class Members for various reasons will not file Claim Forms and that the payments to Eligible Class Members will be made on a claims-made basis."
It would also be a bonanza for plaintiffs' counsel. It contemplates that the company would pay attorney's fees in the amount of $2.25 million regardless of how many claims are actually submitted. The $2.25 million might wind up being more in fees than the class receives in payments. We must remember that the Settlement this is a claims-made settlement, not a common-fund settlement (see note 3 below).
The proposal would be also a bonanza for the named representatives. They would not be stuck with the dimes-on-a-dollar discount negotiated for the class. The three named plaintiffs would each also receive an additional kicker of $15,000 as a so-called "incentive payment" for a total of $45,000.
Without doubt, the main losers under this proposal would be those absent class members who wind up not submitting a timely claim and/or who never receive a notice letter in the first place. All of their rights to seek unpaid wages individually (or through representative actions in other states) would be erased totally. The Settlement Agreement expressly provides that mailing to last known addresses with follow-up tracing on undelivered envelopes would be sufficient to forfeit workers' rights even if they never received the letter, as many would not due to mobility and changes of addresses (¶¶ 20, 23(c)). The Settlement Agreement expressly recognizes that some workers will never get any actual notice and/or submit claims (¶¶ 2(d), 23). Nonetheless, Oracle specifically extracted a concession that all settlement amounts attributable to those workers would revert to Oracle (¶¶ 27(d), 32), yet those workers' rights would be erased (¶ 23(c)).
It is true, under the agreement, that if fewer than seventy percent of all potential claimants turn in claim forms, then the named plaintiffs would have the right to abort the settlement. But the provision expressly does not require it — the named plaintiffs and their counsel would be free to proceed with the proposal even if, for example, only half submitted claims — despite the fact that the other half would lose all rights to sue in the blanket nationwide release. Given the special incentive payments to incentivize the named plaintiffs, this provision is cold comfort.
In sum, the fundamental substance of this proposal is so unfair to absent class members that it cannot pass even the threshold of plausibility required for even preliminary approval under Rule 23. For all workers, all wage-and-hour rights under federal and various state laws and statutory limitations periods, including the various local laws where they reside, would be extinguished and replaced by a fleeting opportunity to prepare and to deliver a claim on a compressed schedule. Workers would be saddled as well with a covenant not to sue that would bar all suits, expressly even existing lawsuits pending elsewhere in the country (¶ 15(b)). The chance that notice by mail would really reach all workers is very remote, especially since many are former workers and have scattered to the winds, yet the release would be absolute and universal. From practical experience, the Court is confident that a significant proportion of all workers would not submit timely claims. No doubt, this is what the company is counting on to reduce its exposure. No doubt, this is also why counsel want a fee based on the theoretical maximum payout rather than the actual benefit conferred.
Precisely the same benefit could have been conferred by making the release coextensive with Oracle's liability, i.e., all those who receive money under the settlement would release their claims — those who do not submit would lose no rights.
3. UNFAIR DISPARITY BETWEEN CALIFORNIA AND NON-CALIFORNIA WORKERS.
As to those who do submit claims, the record does not justify the disparity non-California workers would have to accept on their claims. The settlement calls for multiple subclasses. There is a complicated allocation formula. Although it took the Court several tries to understand how it would work (reading and re-reading the proposed class notice), it finally sunk in that California claimants would receive at least twice as much as non-California claimants. Why a sales consultant in New York should receive half of what an identically situated sales consultant in California would receive for the same overtime has not been explained. No cogent justification for such an allocation has been submitted. This is all the more egregious because the vast majority of the affected workers reside outside California. In California there are 358 affected workers. Outside California there are 1432 affected workers in the various subclasses. This is a warning flag indicating that what started out as a California class action (plus a collective opt-in FLSA action) was seized on by Oracle as a vehicle for a cheap nationwide absolution of its back-pay obligations.
4. INADEQUATE JUSTIFICATION FOR STEEP DISCOUNT.
Assuming for the sake of argument that all workers would receive notice and file claims and that all nine million dollars (less fees) would be paid to claimants, we must compare that amount against the possible recoveries at trial. No evidence whatsoever was submitted with the motion to explain the extent of possible recoveries. The Court then requested a copy of the plaintiffs' expert damages report. What was submitted was not a sworn expert report but unsworn spreadsheets prepared by counsel. The maximum recovery according to the material would be $52.7 million. This means that the settlement would, on average yield a 12.3 percent recovery (after fees) but only for those submitting claims. In other words, claimants would forfeit 87.7 percent of their maximum claim. In essence, the reason given for this steep discount is that counsel feel it is "imminently reasonable in light of the risks inherent in this, as any, wage and hour class action" (Willson Decl. ¶ 3). Perhaps this is so. Perhaps not. While this Court would defer to the judgment of counsel in balancing the risks versus rewards of litigation, more must be explained than so far has been to justify the steep discount. This is curable but the record must support the steep discount. In doing so, what matters are the challenges and likelihoods of success on the merits of the substantive statutes at issue, not the odds that a class would be certified.
5. FLSA.
The settlement agreement would violate the Federal Fair Labor Standards Act. The FLSA prohibits traditional class actions and authorizes only an opt-in collective action. 29 U.S.C. 216(b). Under no circumstances can counsel collude to take away FLSA rights including the worker's right to control his or her own claim without the burden of having to opt out of someone else's lawsuit. Workers who voluntarily send in a claim form and affirmatively join in the action, of course, can be bound to a full release of all federal and state rights. But is unconscionable to try to take away the FLSA rights of all workers, whether or not they choose to join in affirmatively.
In response to the Court's questions, counsel have pretended that the instruments would only release FLSA claims of those who opt in. For the reasons stated above, the actual documentation is to the contrary. To take just one example, the proposed notice states: "Eligible Class Members waive and release their proposed right to sue with respect to any claim under the Fair Labor Standards Act. . . . "
In response to the Court's questions regarding the proposal, inconsistent answers were given to questions concerning the FLSA, among others. Response No. 3 states that FLSA claims would be waived for opt-ins but only for non-California class members. This implies that California opt-ins would not waive their FLSA claims, which would be odd and unfair. Presumably, all opt-ins should waive all wage-and-hour claims under any and all laws.
6. JURISDICTION OVER NON-CALIFORNIA STATE CLAIMS OF NON-CALIFORNIA WORKERS.
There is a serious problem of personal jurisdiction over non-California workers and the release of state-law claims outside California. Once again, those non-residents who actually do submit claim forms may voluntarily agree to submit to the jurisdiction of the Court and to waive all claims, California and non-California (as well as FLSA). But for those in, say, New York or Mississippi who receive a notice letter from California and do not file a claim form, a serious question exists whether they can be forced to submit to a California court's jurisdiction and be bound by an order extinguishing their rights.
As a matter of personal jurisdiction, the Supreme Court has allowed a state court to assert jurisdiction over class members receiving actual notice, at least in the circumstances of that particular case. Philips Petroleum Co. v. Shutts, 472 U.S. 797, 806-14 (1985) (undelivered letters automatically excluded from class). Despite an explicit request to do so, however, counsel have cited no authority holding that a nationwide class under Section 17200 can be used to extinguish wage-and-hour claims under non-California law for non-California residents not submitting claim forms.
Even when a nationwide class is otherwise appropriate, a court in California cannot automatically apply California law to adjudicate the wage-and-hour claims of workers in New York or Mississippi or elsewhere. "The issue of personal jurisdiction over plaintiffs on a class action is entirely distinct from the question of the constitutional limitations on choice of law; the latter calculus is not altered by the fact that it may be more difficult or more burdensome to comply with the constitutional limitations because of the large number of transactions which the State proposes to adjudicate and which have little connection to the forum." Phillips Petroleum Co., 472 U.S. 821. As to each absent class member, a choice-of-law analysis must be done. It may well be, for example, that the choice-of-law analysis would indicate that New York law should govern the rights of a New York worker who put in overtime at a New York facility. None of this analysis has been provided to address the 150-plus laws of 35 other states that Oracle would extinguish under cover of Section 17200. Nor are our three named plaintiffs representative of the 35 other states (except possibly Washington).
If the release were narrowed to apply only to those receiving money under the settlement, then, as stated, all federal and state laws anywhere could be included in the release. But as now written, this is not the case.
7. LACK OF ADEQUATE REPRESENTATION.
The named plaintiffs worked in California and Washington state, the latter purporting to sue under California law. They are not adequate to represent residents of, say, New York and Mississippi. Workers' rights vary state by state. Special protections, penalties and limitation periods may exist in some states that do not exist in California. This is illustrated in high relief by the list of 150 various laws of 35 states provided by Oracle that it contends would be extinguished by the proposal. The named plaintiffs are not in a position to assess the state and local rights in various corners of the country.
A different problem of adequacy of representation is that there are four subclasses but only three plaintiffs. There is no named plaintiff for Subclass No. 2, the class covering the time period from September 20, 2004, forward. The same is true for the parallel purported FLSA collective subclass.
Yet, a further representation problem is that while the plaintiffs' claim is for unpaid overtime, the classes are to waive any and all wage-and-hour claims, not just overtime claims. No explanation has been provided how these three named plaintiffs are adequate to represent the much broader set of claims.
It is also worth noting that the record contains no declarations from the three plaintiffs and insufficient information to verify that they should be trusted with the fiduciary roles at issue.
8. ATTORNEY'S FEES.
It is premature now to bless any particular fee award, even on a preliminary basis. A proper motion supported by time records and all other necessary evidence would be needed, along with an opportunity to evaluate objections. Although counsel desire a percentage recovery, the Ninth Circuit requires the district judge to take time records and other facts into account before using a percentage method. Six Mexican Workers v. Ariz. Citrus Growers, 904 F.2d 1301, 1311 (9th Cir. 1990).
Nevertheless, some observations may be helpful. Contrary to counsel's rhetoric and unlike the decisions cited by counsel, this is not a common fund case. The Settlement Agreement pointedly disavows any common fund, a disavowal intended expressly to defeat any application of Section 384 of the California Code of Civil Procedure. The Agreement states (¶ 33) (emphasis added):
The parties agree that this Settlement Agreement does not create or establish a settlement or common "fund," and that all claims for payment shall be submitted directly to and paid by the Claims Administrator. The parties represent that the Funds Available for Settlement is a settlement amount that takes into account the probability that some Eligible Class Members for various reasons will not file Claim Forms and that the payments to Eligible Class Members will be made on a claims-made basis.
Therefore, Oracle will only have to pay out "on a claims-made basis." All amounts not paid to class members revert to Oracle, including interest. Not one cent will be available for cy pres.
A $2.25 million fee to counsel might (or might not) wind up being excessive under Six Mexican Workers. The Settlement Agreement acknowledges "[t]he probability that some Eligible Class Members for various reasons will not file Claim Forms and that the payments to Eligible Class Members will be made on a claims-made basis" (¶ 33). Those for whom no claim is filed would lose all rights, even if they never get actual notice. To grant attorney's fees based on claims not submitted might wind up being perverse, for it would, as to non-claimants, reward counsel for achieving nothing more than a forfeiture of statutory and regulatory rights, at least as the instruments are worded at present.
Again, it is premature to approve, even preliminarily, any particular amount of fees or approach to granting fees. Whatever the claims procedure might be, assuming all other problems could be solved, it must allow for a final decision on fees before final checks are to be cut for claimants, as their recovery could go up or down as a result.
9. INCENTIVE PAYMENTS.
So-called "incentive payments" are a recent invention by those who handle class actions. Class actions did much justice without them for many decades. While there is a theoretical rationale for incentive payments, there is also a major downside. The downside is that the payments lend themselves for use as side payments to induce named plaintiffs to go along with sweetheart deals. Ordinarily, named plaintiffs ought to receive no more or less than the absent class members they purport to represent. In this way, they are incentivized, out of self interest, to achieve the best possible result for the class. The Court finds that the record does not justify the proposed extra payments.
For the foregoing reasons, the proposal at issue gives the unfortunate appearance that those in charge of this case have been co-opted at the expense of the absent workers and that a California class suit has been enlarged simply to compromise workers on a national scale and to roll in all wage-and-hour claims, not just overtime claims, so as to magnify fees.
10. METHOD OF NOTICE.
In the circumstances of this case, notice by mail to last known addresses would not be adequate. For starters, most of the absent class workers still work for Oracle. In addition to mail delivery, it would be very effective and highly practicable to provide these current employees with workplace notice, either by hard copy or e-mail. Mail alone is not the best practicable notice.
As to former employees, any actual service by first-class mail (not certified mail) to a correct address would be satisfactory. Service, however, on the "last known address" has not yet been shown to be good enough. There are former employees whose employment goes back to 2000. One named plaintiff now lives in Europe. These former workers are now scattered to the winds. We must remember that Oracle is trying to use a California court to erase the workers' rights in 35 states and under 150 different local laws. The present record does not reveal the accuracy rate of the proposed procedure. Nor does it identify alternatives, such as publication and/or the Internet, and explain why they are unacceptable as supplemental modes of notice. As to former workers, it may be necessary to wait and see how effective the delivery rate turns out to be before blessing or not blessing the procedure.
Notice by first-class mail has been held sufficient by the Supreme Court where the mail is delivered and where those that could not have delivered were excluded from the class. Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 812-13 (1985); Ortiz v. Fibreboard Corp., 527 U.S. 815, 848 (1999). This is so that the absent class members can remove themselves from the class. Some caselaw allows mailed notice to be deemed adequate even when it is not delivered correctly but there must be a showing that the proposed address list is reasonably accurate. Otherwise, notice by publication may be necessary.
11. INCOMPREHENSIBILITY OF THE PROPOSED NOTICE.
The proposal would break the workers down into four subclasses. One would cover California residents in Time Period One. Another would cover California residents in Time Period Two. Another would cover non-California residents in Time Period Three and another would cover non-California residents in Time Period Four. Then there would be a parallel set of four further subclasses defined in the same way but designated as for "provisional certification of this action as an FLSA collective action." So there would be eight subclasses. Every class member must be in at least two, but some could be in more than two. This convoluted tangle of subclasses is, it seems to the Court, driven by a collusive desire to cosmetically gloss over the jurisdictional issues above, namely the attempt to somehow fold in non-California claims as well as FLSA claims. This contrived structure is simply too hard to understand. At least as proposed, the notice does a poor job of explaining the subclasses and the benefits and disadvantages of the proposal.
The notice is deficient in other ways as well. For example, it uses a defined term "Eligible Class Member," but the notice itself does not define it. True, the settlement agreement defines the term, but class members will not receive that document. Nor will they receive the list of 150 statutes in 35 states appended as an exhibit to the Settlement Agreement. Defined terms and legalese and cascading incorporations by reference are confusing and should be sparingly used. Plain English is needed. Other drafting errors could be listed but this order will end here. The proposal being so beyond the pale that flyspecking the notice would be pointless.
12. THE RIGHT TO OPT OUT IS NOT A CURE-ALL.
While it is true that workers who dislike the proposal could opt out, this does not absolve the Court from its independent duty to vet the proposal. Nor does the fact that workers would have the right to submit objections. It is worth reminding ourselves that few ever opt out or object. Those who file claim forms do not appreciate the questions of substance and fairness that will suggest themselves to a district judge. There is no substitute for the requirement of district courts vetting the proposed settlement under Rule 23(e).
It is also no answer to say that a private mediator helped frame the proposal. Such a mediator is paid to help the immediate parties reach a deal. Mediators do not adjudicate the merits. They are masters in the art of what is negotiable. It matters little to the mediator whether a deal is collusive as long as a deal is reached. Such a mediator has no fiduciary duty to anyone, much less those not at the table. Plaintiffs' counsel has the fiduciary duty. It cannot be delegated to a private mediator.
* * *
Virtually none of the foregoing problems were raised by counsel, illustrating the sad fact that once a collusive settlement is reached, counsel have no incentive to critique their joint proposal. The district judge must dig through the file on his or her own. No doubt, this Court has missed some further issues not having an advocate to call out the weak spots.
CONCLUSION
A single modification would go a long way (but not all the way) forward curing the foregoing problems, namely to provide that the release would only be effective against those submitting claims, i.e., opting into the settlement. Put differently, since Oracle's liability is expressly placed on a "claims-made basis (¶ 33), so too the release might be imposed only on a "claims-made" basis. This would probably require a renegotiation of other terms as well, including the economics. The Court, of course, cannot force an unwanted settlement on anyone. So, whether the parties wish to try again is up to them. Meanwhile, the case will proceed on schedule.Preliminary approval is DENIED. The motion for class certification filed earlier shall be heard on AUGUST 2 AT 8:00 A.M. As extended, the opposition is JULY 12, and the reply is due JULY 19, AT NOON IN BOTH CASES. Tentatively, JULY 5 AT 3:00 P.M. will be reserved to consider any new attempt at class settlement. The parties are requested to submit chambers copies of both redacted and unredacted versions of the motion for class certification and the accompanying declarations.